{ "generated_at": "2026-10-24T20:00:00Z", "asset_type": "skill", "count": 56, "items": [ { "id": "skl_8f2d1534b122", "slug": "skill-workers-best-practices", "asset_type": "skill", "name": "Cloudflare Workers Production Architecture Playbook", "tagline": "Production engineering standards for Cloudflare Workers: V8 isolate memory limits, streaming Fetch responses, bindings security, CPU budgets, and observabi", "description_md": "### Overview & Capability Scope\n**Cloudflare Workers Production Architecture Playbook** (`skill-workers-best-practices`) is a verified, portable `SKILL.md` playbook curated with **@cloudflare** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nProduction engineering standards for Cloudflare Workers: V8 isolate memory limits, streaming Fetch responses, bindings security, CPU budgets, and observability.\n\n### Required Stack & Tooling\n- **Primary Stack:** wrangler.jsonc, Cloudflare Workers, TypeScript, Observability\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/cloudflare/skills](https://github.com/cloudflare/skills)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-workers-best-practices\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-workers-best-practices/SKILL.md` or `~/.claude/skills/skill-workers-best-practices/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx wrangler dev\nnpx wrangler deploy --dry-run\nnpx wrangler tail --format=json\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "wrangler.jsonc", "cloudflare-workers", "typescript", "observability", "skill-md", "engineering" ], "source_url": "https://github.com/cloudflare/skills", "repo_url": "https://github.com/cloudflare/skills", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-workers-best-practices", "skill_markdown": "---\nname: workers-best-practices\ndescription: Production engineering standards for Cloudflare Workers: V8 isolate memory limits, streaming Fetch responses, bindings security, CPU budgets, and observability. Use when working with wrangler.jsonc, Cloudflare Workers, TypeScript, Observability.\ntriggers:\n - workers best practices\n - wrangler.jsonc\n - cloudflare workers\n---\n\n# Cloudflare Workers Production Architecture Playbook\n\n## 1. When to Use This Skill\nLoad `skill-workers-best-practices` whenever an engineering or operations task involves **wrangler.jsonc, Cloudflare Workers, TypeScript, Observability**. Specifically:\n- Production engineering standards for Cloudflare Workers: V8 isolate memory limits, streaming Fetch responses, bindings security, CPU budgets, and observability.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** wrangler.jsonc, Cloudflare Workers, TypeScript, Observability\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/cloudflare/skills`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **wrangler.jsonc**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx wrangler dev\nnpx wrangler deploy --dry-run\nnpx wrangler tail --format=json\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Cloudflare Workers Production Architecture Playbook (skill-workers-best-practices) playbook. First inspect our current configuration and codebase for wrangler.jsonc, Cloudflare Workers, TypeScript, Observability. Then propose a concrete, step-by-step implementation plan following the skill-workers-best-practices guardrails, run the non-destructive verification commands (npx wrangler dev), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "wrangler.jsonc", "Cloudflare Workers", "TypeScript" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "cloudflare", "creator_name": "Cloudflare Developers", "creator_url": "https://x.com/cloudflare", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 94, "installs_count": 420, "agent_pulls_count": 890, "gravity_score": 3295.0, "is_featured": true, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_da8c5f72ae77", "slug": "skill-wrangler", "asset_type": "skill", "name": "Wrangler CLI Deployment & Resource Management Playbook", "tagline": "Deterministic workflow for configuring `wrangler.jsonc`, provisioning D1/KV/R2 bindings, managing secrets, and running preview vs production deployments.", "description_md": "### Overview & Capability Scope\n**Wrangler CLI Deployment & Resource Management Playbook** (`skill-wrangler`) is a verified, portable `SKILL.md` playbook curated with **@cloudflare** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nDeterministic workflow for configuring `wrangler.jsonc`, provisioning D1/KV/R2 bindings, managing secrets, and running preview vs production deployments.\n\n### Required Stack & Tooling\n- **Primary Stack:** Wrangler CLI, Cloudflare Workers, D1, KV, R2\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/cloudflare/skills](https://github.com/cloudflare/skills)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-wrangler\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-wrangler/SKILL.md` or `~/.claude/skills/skill-wrangler/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx wrangler whoami\nnpx wrangler d1 list\nnpx wrangler secret put API_KEY\nnpx wrangler deploy\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "wrangler-cli", "cloudflare-workers", "d1", "kv", "skill-md", "engineering" ], "source_url": "https://github.com/cloudflare/skills", "repo_url": "https://github.com/cloudflare/skills", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-wrangler", "skill_markdown": "---\nname: wrangler\ndescription: Deterministic workflow for configuring `wrangler.jsonc`, provisioning D1/KV/R2 bindings, managing secrets, and running preview vs production deployments. Use when working with Wrangler CLI, Cloudflare Workers, D1, KV, R2.\ntriggers:\n - wrangler\n - wrangler cli\n - cloudflare workers\n---\n\n# Wrangler CLI Deployment & Resource Management Playbook\n\n## 1. When to Use This Skill\nLoad `skill-wrangler` whenever an engineering or operations task involves **Wrangler CLI, Cloudflare Workers, D1, KV, R2**. Specifically:\n- Deterministic workflow for configuring `wrangler.jsonc`, provisioning D1/KV/R2 bindings, managing secrets, and running preview vs production deployments.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Wrangler CLI, Cloudflare Workers, D1, KV, R2\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/cloudflare/skills`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Wrangler CLI**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx wrangler whoami\nnpx wrangler d1 list\nnpx wrangler secret put API_KEY\nnpx wrangler deploy\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Wrangler CLI Deployment & Resource Management Playbook (skill-wrangler) playbook. First inspect our current configuration and codebase for Wrangler CLI, Cloudflare Workers, D1, KV, R2. Then propose a concrete, step-by-step implementation plan following the skill-wrangler guardrails, run the non-destructive verification commands (npx wrangler whoami), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Wrangler CLI", "Cloudflare Workers", "D1" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "cloudflare", "creator_name": "Cloudflare Developers", "creator_url": "https://x.com/cloudflare", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 93, "installs_count": 416, "agent_pulls_count": 883, "gravity_score": 3266.0, "is_featured": true, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_63cff570d320", "slug": "skill-cloudflare-d1-migrations", "asset_type": "skill", "name": "Cloudflare D1 SQLite Schema & Zero-Downtime Migrations Skill", "tagline": "Design indexed D1 SQLite schemas, author forward-only SQL migrations, benchmark query latency with `PRAGMA optimize`, and sync local and remote D1 database", "description_md": "### Overview & Capability Scope\n**Cloudflare D1 SQLite Schema & Zero-Downtime Migrations Skill** (`skill-cloudflare-d1-migrations`) is a verified, portable `SKILL.md` playbook curated with **@cloudflare** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nDesign indexed D1 SQLite schemas, author forward-only SQL migrations, benchmark query latency with `PRAGMA optimize`, and sync local and remote D1 databases.\n\n### Required Stack & Tooling\n- **Primary Stack:** Cloudflare D1, SQLite, Wrangler CLI, SQL Migrations\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/cloudflare/skills](https://github.com/cloudflare/skills)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-cloudflare-d1-migrations\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-cloudflare-d1-migrations/SKILL.md` or `~/.claude/skills/skill-cloudflare-d1-migrations/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx wrangler d1 migrations create db init_schema\nnpx wrangler d1 migrations apply db --local\nnpx wrangler d1 migrations apply db --remote\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "cloudflare-d1", "sqlite", "wrangler-cli", "sql-migrations", "skill-md", "engineering" ], "source_url": "https://github.com/cloudflare/skills", "repo_url": "https://github.com/cloudflare/skills", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-cloudflare-d1-migrations", "skill_markdown": "---\nname: cloudflare-d1-migrations\ndescription: Design indexed D1 SQLite schemas, author forward-only SQL migrations, benchmark query latency with `PRAGMA optimize`, and sync local and remote D1 databases. Use when working with Cloudflare D1, SQLite, Wrangler CLI, SQL Migrations.\ntriggers:\n - cloudflare d1 migrations\n - cloudflare d1\n - sqlite\n---\n\n# Cloudflare D1 SQLite Schema & Zero-Downtime Migrations Skill\n\n## 1. When to Use This Skill\nLoad `skill-cloudflare-d1-migrations` whenever an engineering or operations task involves **Cloudflare D1, SQLite, Wrangler CLI, SQL Migrations**. Specifically:\n- Design indexed D1 SQLite schemas, author forward-only SQL migrations, benchmark query latency with `PRAGMA optimize`, and sync local and remote D1 databases.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Cloudflare D1, SQLite, Wrangler CLI, SQL Migrations\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/cloudflare/skills`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Cloudflare D1**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx wrangler d1 migrations create db init_schema\nnpx wrangler d1 migrations apply db --local\nnpx wrangler d1 migrations apply db --remote\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Cloudflare D1 SQLite Schema & Zero-Downtime Migrations Skill (skill-cloudflare-d1-migrations) playbook. First inspect our current configuration and codebase for Cloudflare D1, SQLite, Wrangler CLI, SQL Migrations. Then propose a concrete, step-by-step implementation plan following the skill-cloudflare-d1-migrations guardrails, run the non-destructive verification commands (npx wrangler d1 migrations create db init_schema), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Cloudflare D1", "SQLite", "Wrangler CLI" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "cloudflare", "creator_name": "Cloudflare Developers", "creator_url": "https://x.com/cloudflare", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 92, "installs_count": 412, "agent_pulls_count": 876, "gravity_score": 3237.0, "is_featured": true, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_baf3fabc2e53", "slug": "skill-cloudflare-r2-storage", "asset_type": "skill", "name": "Cloudflare R2 Object Storage & Presigned URL Pipeline Skill", "tagline": "Zero-egress object storage architecture on Cloudflare R2: S3-compatible presigned upload URLs, multipart streaming, CORS policies, and custom domain cachin", "description_md": "### Overview & Capability Scope\n**Cloudflare R2 Object Storage & Presigned URL Pipeline Skill** (`skill-cloudflare-r2-storage`) is a verified, portable `SKILL.md` playbook curated with **@cloudflare** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nZero-egress object storage architecture on Cloudflare R2: S3-compatible presigned upload URLs, multipart streaming, CORS policies, and custom domain caching.\n\n### Required Stack & Tooling\n- **Primary Stack:** Cloudflare R2, S3 API, Workers, Presigned URLs\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/cloudflare/skills](https://github.com/cloudflare/skills)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-cloudflare-r2-storage\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-cloudflare-r2-storage/SKILL.md` or `~/.claude/skills/skill-cloudflare-r2-storage/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx wrangler r2 bucket create assets-prod\nnpx wrangler r2 object put assets-prod/index.json --file=./data.json\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "cloudflare-r2", "s3-api", "workers", "presigned-urls", "skill-md", "engineering" ], "source_url": "https://github.com/cloudflare/skills", "repo_url": "https://github.com/cloudflare/skills", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-cloudflare-r2-storage", "skill_markdown": "---\nname: cloudflare-r2-storage\ndescription: Zero-egress object storage architecture on Cloudflare R2: S3-compatible presigned upload URLs, multipart streaming, CORS policies, and custom domain caching. Use when working with Cloudflare R2, S3 API, Workers, Presigned URLs.\ntriggers:\n - cloudflare r2 storage\n - cloudflare r2\n - s3 api\n---\n\n# Cloudflare R2 Object Storage & Presigned URL Pipeline Skill\n\n## 1. When to Use This Skill\nLoad `skill-cloudflare-r2-storage` whenever an engineering or operations task involves **Cloudflare R2, S3 API, Workers, Presigned URLs**. Specifically:\n- Zero-egress object storage architecture on Cloudflare R2: S3-compatible presigned upload URLs, multipart streaming, CORS policies, and custom domain caching.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Cloudflare R2, S3 API, Workers, Presigned URLs\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/cloudflare/skills`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Cloudflare R2**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx wrangler r2 bucket create assets-prod\nnpx wrangler r2 object put assets-prod/index.json --file=./data.json\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Cloudflare R2 Object Storage & Presigned URL Pipeline Skill (skill-cloudflare-r2-storage) playbook. First inspect our current configuration and codebase for Cloudflare R2, S3 API, Workers, Presigned URLs. Then propose a concrete, step-by-step implementation plan following the skill-cloudflare-r2-storage guardrails, run the non-destructive verification commands (npx wrangler r2 bucket create assets-prod), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Cloudflare R2", "S3 API", "Workers" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "cloudflare", "creator_name": "Cloudflare Developers", "creator_url": "https://x.com/cloudflare", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 91, "installs_count": 408, "agent_pulls_count": 869, "gravity_score": 3208.0, "is_featured": true, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_cf227a67dbf9", "slug": "skill-durable-objects", "asset_type": "skill", "name": "Cloudflare Durable Objects Stateful Coordination Playbook", "tagline": "Build strongly consistent, single-threaded coordination actors with Cloudflare Durable Objects, SQLite storage backend, WebSockets hibernation, and alarms.", "description_md": "### Overview & Capability Scope\n**Cloudflare Durable Objects Stateful Coordination Playbook** (`skill-durable-objects`) is a verified, portable `SKILL.md` playbook curated with **@threepointone** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nBuild strongly consistent, single-threaded coordination actors with Cloudflare Durable Objects, SQLite storage backend, WebSockets hibernation, and alarms.\n\n### Required Stack & Tooling\n- **Primary Stack:** Durable Objects, WebSockets, SQLite Storage, Cloudflare Workers\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/cloudflare/skills](https://github.com/cloudflare/skills)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-durable-objects\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-durable-objects/SKILL.md` or `~/.claude/skills/skill-durable-objects/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx wrangler dev --test-scheduled\nnpx wrangler deploy\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "durable-objects", "websockets", "sqlite-storage", "cloudflare-workers", "skill-md", "engineering" ], "source_url": "https://github.com/cloudflare/skills", "repo_url": "https://github.com/cloudflare/skills", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-durable-objects", "skill_markdown": "---\nname: durable-objects\ndescription: Build strongly consistent, single-threaded coordination actors with Cloudflare Durable Objects, SQLite storage backend, WebSockets hibernation, and alarms. Use when working with Durable Objects, WebSockets, SQLite Storage, Cloudflare Workers.\ntriggers:\n - durable objects\n - durable objects\n - websockets\n---\n\n# Cloudflare Durable Objects Stateful Coordination Playbook\n\n## 1. When to Use This Skill\nLoad `skill-durable-objects` whenever an engineering or operations task involves **Durable Objects, WebSockets, SQLite Storage, Cloudflare Workers**. Specifically:\n- Build strongly consistent, single-threaded coordination actors with Cloudflare Durable Objects, SQLite storage backend, WebSockets hibernation, and alarms.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Durable Objects, WebSockets, SQLite Storage, Cloudflare Workers\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/cloudflare/skills`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Durable Objects**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx wrangler dev --test-scheduled\nnpx wrangler deploy\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Cloudflare Durable Objects Stateful Coordination Playbook (skill-durable-objects) playbook. First inspect our current configuration and codebase for Durable Objects, WebSockets, SQLite Storage, Cloudflare Workers. Then propose a concrete, step-by-step implementation plan following the skill-durable-objects guardrails, run the non-destructive verification commands (npx wrangler dev --test-scheduled), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Durable Objects", "WebSockets", "SQLite Storage" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "threepointone", "creator_name": "Sunil Pai (@threepointone)", "creator_url": "https://x.com/threepointone", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 90, "installs_count": 404, "agent_pulls_count": 862, "gravity_score": 3179.0, "is_featured": true, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_e52bed00cf09", "slug": "skill-agents-sdk", "asset_type": "skill", "name": "Cloudflare Agents SDK Stateful AI Runtime Playbook", "tagline": "Build, debug, and deploy persistent AI agents on Cloudflare using the `agents` SDK (`Agent` class, `useAgent` client hooks, scheduled tasks, and MCP client", "description_md": "### Overview & Capability Scope\n**Cloudflare Agents SDK Stateful AI Runtime Playbook** (`skill-agents-sdk`) is a verified, portable `SKILL.md` playbook curated with **@threepointone** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nBuild, debug, and deploy persistent AI agents on Cloudflare using the `agents` SDK (`Agent` class, `useAgent` client hooks, scheduled tasks, and MCP client bindings).\n\n### Required Stack & Tooling\n- **Primary Stack:** @cloudflare/agents, Durable Objects, Streamable HTTP MCP, Workers AI\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/cloudflare/agents](https://github.com/cloudflare/agents)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-agents-sdk\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-agents-sdk/SKILL.md` or `~/.claude/skills/skill-agents-sdk/SKILL.md`.\n\n### Verification Commands\n```bash\nnpm i agents\nnpx wrangler deploy\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "@cloudflare/agents", "durable-objects", "streamable-http-mcp", "workers-ai", "skill-md", "engineering" ], "source_url": "https://github.com/cloudflare/agents", "repo_url": "https://github.com/cloudflare/agents", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-agents-sdk", "skill_markdown": "---\nname: agents-sdk\ndescription: Build, debug, and deploy persistent AI agents on Cloudflare using the `agents` SDK (`Agent` class, `useAgent` client hooks, scheduled tasks, and MCP client bindings). Use when working with @cloudflare/agents, Durable Objects, Streamable HTTP MCP, Workers AI.\ntriggers:\n - agents sdk\n - @cloudflare/agents\n - durable objects\n---\n\n# Cloudflare Agents SDK Stateful AI Runtime Playbook\n\n## 1. When to Use This Skill\nLoad `skill-agents-sdk` whenever an engineering or operations task involves **@cloudflare/agents, Durable Objects, Streamable HTTP MCP, Workers AI**. Specifically:\n- Build, debug, and deploy persistent AI agents on Cloudflare using the `agents` SDK (`Agent` class, `useAgent` client hooks, scheduled tasks, and MCP client bindings).\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** @cloudflare/agents, Durable Objects, Streamable HTTP MCP, Workers AI\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/cloudflare/agents`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **@cloudflare/agents**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpm i agents\nnpx wrangler deploy\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Cloudflare Agents SDK Stateful AI Runtime Playbook (skill-agents-sdk) playbook. First inspect our current configuration and codebase for @cloudflare/agents, Durable Objects, Streamable HTTP MCP, Workers AI. Then propose a concrete, step-by-step implementation plan following the skill-agents-sdk guardrails, run the non-destructive verification commands (npm i agents), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "@cloudflare/agents", "Durable Objects", "Streamable HTTP MCP" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "threepointone", "creator_name": "Sunil Pai (@threepointone)", "creator_url": "https://x.com/threepointone", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 89, "installs_count": 400, "agent_pulls_count": 855, "gravity_score": 3150.0, "is_featured": true, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_3f7455b7aaba", "slug": "skill-turnstile-spin", "asset_type": "skill", "name": "Cloudflare Turnstile Bot Verification & Siteverify Skill", "tagline": "Implement privacy-preserving Cloudflare Turnstile challenge widgets on frontend forms with mandatory server-side Siteverify token validation and replay pro", "description_md": "### Overview & Capability Scope\n**Cloudflare Turnstile Bot Verification & Siteverify Skill** (`skill-turnstile-spin`) is a verified, portable `SKILL.md` playbook curated with **@cloudflare** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nImplement privacy-preserving Cloudflare Turnstile challenge widgets on frontend forms with mandatory server-side Siteverify token validation and replay protection.\n\n### Required Stack & Tooling\n- **Primary Stack:** Cloudflare Turnstile, Siteverify API, Edge Security, Form Protection\n- **Category:** `security-ops`\n- **Source & Reference:** [https://github.com/cloudflare/skills](https://github.com/cloudflare/skills)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-turnstile-spin\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-turnstile-spin/SKILL.md` or `~/.claude/skills/skill-turnstile-spin/SKILL.md`.\n\n### Verification Commands\n```bash\ncurl -s https://challenges.cloudflare.com/turnstile/v0/siteverify -d secret=$TURNSTILE_SECRET -d response=$TOKEN\n```\n", "category": "security-ops", "subcategories": [ "security-ops", "skills" ], "tags": [ "cloudflare-turnstile", "siteverify-api", "edge-security", "form-protection", "skill-md", "security-ops" ], "source_url": "https://github.com/cloudflare/skills", "repo_url": "https://github.com/cloudflare/skills", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-turnstile-spin", "skill_markdown": "---\nname: turnstile-spin\ndescription: Implement privacy-preserving Cloudflare Turnstile challenge widgets on frontend forms with mandatory server-side Siteverify token validation and replay protection. Use when working with Cloudflare Turnstile, Siteverify API, Edge Security, Form Protection.\ntriggers:\n - turnstile spin\n - cloudflare turnstile\n - siteverify api\n---\n\n# Cloudflare Turnstile Bot Verification & Siteverify Skill\n\n## 1. When to Use This Skill\nLoad `skill-turnstile-spin` whenever an engineering or operations task involves **Cloudflare Turnstile, Siteverify API, Edge Security, Form Protection**. Specifically:\n- Implement privacy-preserving Cloudflare Turnstile challenge widgets on frontend forms with mandatory server-side Siteverify token validation and replay protection.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Cloudflare Turnstile, Siteverify API, Edge Security, Form Protection\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/cloudflare/skills`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Cloudflare Turnstile**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\ncurl -s https://challenges.cloudflare.com/turnstile/v0/siteverify -d secret=$TURNSTILE_SECRET -d response=$TOKEN\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Cloudflare Turnstile Bot Verification & Siteverify Skill (skill-turnstile-spin) playbook. First inspect our current configuration and codebase for Cloudflare Turnstile, Siteverify API, Edge Security, Form Protection. Then propose a concrete, step-by-step implementation plan following the skill-turnstile-spin guardrails, run the non-destructive verification commands (curl -s https://challenges.cloudflare.com/turnstile/v0/siteverify -d secret=$TURNSTILE_SECRET -d response=$TOKEN), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Cloudflare Turnstile", "Siteverify API", "Edge Security" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "cloudflare", "creator_name": "Cloudflare Developers", "creator_url": "https://x.com/cloudflare", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 88, "installs_count": 396, "agent_pulls_count": 848, "gravity_score": 3121.0, "is_featured": true, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_a0d1c095dd1d", "slug": "skill-nextjs-on-cloudflare", "asset_type": "skill", "name": "Next.js on Cloudflare Workers & OpenNext Migration Playbook", "tagline": "Deploy full-stack Next.js App Router applications to Cloudflare Workers using OpenNext and vinext with edge caching, ISR via R2/KV, and bindings integratio", "description_md": "### Overview & Capability Scope\n**Next.js on Cloudflare Workers & OpenNext Migration Playbook** (`skill-nextjs-on-cloudflare`) is a verified, portable `SKILL.md` playbook curated with **@cloudflare** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nDeploy full-stack Next.js App Router applications to Cloudflare Workers using OpenNext and vinext with edge caching, ISR via R2/KV, and bindings integration.\n\n### Required Stack & Tooling\n- **Primary Stack:** Next.js, OpenNext, Cloudflare Workers, SSR\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/cloudflare/skills](https://github.com/cloudflare/skills)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-nextjs-on-cloudflare\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-nextjs-on-cloudflare/SKILL.md` or `~/.claude/skills/skill-nextjs-on-cloudflare/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx @opennextjs/cloudflare build\nnpx wrangler deploy\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "next.js", "opennext", "cloudflare-workers", "ssr", "skill-md", "engineering" ], "source_url": "https://github.com/cloudflare/skills", "repo_url": "https://github.com/cloudflare/skills", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-nextjs-on-cloudflare", "skill_markdown": "---\nname: nextjs-on-cloudflare\ndescription: Deploy full-stack Next.js App Router applications to Cloudflare Workers using OpenNext and vinext with edge caching, ISR via R2/KV, and bindings integration. Use when working with Next.js, OpenNext, Cloudflare Workers, SSR.\ntriggers:\n - nextjs on cloudflare\n - next.js\n - opennext\n---\n\n# Next.js on Cloudflare Workers & OpenNext Migration Playbook\n\n## 1. When to Use This Skill\nLoad `skill-nextjs-on-cloudflare` whenever an engineering or operations task involves **Next.js, OpenNext, Cloudflare Workers, SSR**. Specifically:\n- Deploy full-stack Next.js App Router applications to Cloudflare Workers using OpenNext and vinext with edge caching, ISR via R2/KV, and bindings integration.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Next.js, OpenNext, Cloudflare Workers, SSR\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/cloudflare/skills`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Next.js**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx @opennextjs/cloudflare build\nnpx wrangler deploy\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Next.js on Cloudflare Workers & OpenNext Migration Playbook (skill-nextjs-on-cloudflare) playbook. First inspect our current configuration and codebase for Next.js, OpenNext, Cloudflare Workers, SSR. Then propose a concrete, step-by-step implementation plan following the skill-nextjs-on-cloudflare guardrails, run the non-destructive verification commands (npx @opennextjs/cloudflare build), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Next.js", "OpenNext", "Cloudflare Workers" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "cloudflare", "creator_name": "Cloudflare Developers", "creator_url": "https://x.com/cloudflare", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 87, "installs_count": 392, "agent_pulls_count": 841, "gravity_score": 3092.0, "is_featured": true, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_d16af8f18987", "slug": "skill-cloudflare-one", "asset_type": "skill", "name": "Cloudflare One Zero Trust & Access Policy Architecture Skill", "tagline": "Design and audit Cloudflare Zero Trust Access applications, WARP device posture checks, identity provider OIDC claims, and Cloudflare Tunnel (`cloudflared`", "description_md": "### Overview & Capability Scope\n**Cloudflare One Zero Trust & Access Policy Architecture Skill** (`skill-cloudflare-one`) is a verified, portable `SKILL.md` playbook curated with **@cloudflare** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nDesign and audit Cloudflare Zero Trust Access applications, WARP device posture checks, identity provider OIDC claims, and Cloudflare Tunnel (`cloudflared`) routing.\n\n### Required Stack & Tooling\n- **Primary Stack:** Cloudflare Access, Zero Trust, cloudflared, OIDC\n- **Category:** `security-ops`\n- **Source & Reference:** [https://github.com/cloudflare/skills](https://github.com/cloudflare/skills)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-cloudflare-one\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-cloudflare-one/SKILL.md` or `~/.claude/skills/skill-cloudflare-one/SKILL.md`.\n\n### Verification Commands\n```bash\ncloudflared tunnel list\ncloudflared tunnel route dns app.example.com\n```\n", "category": "security-ops", "subcategories": [ "security-ops", "skills" ], "tags": [ "cloudflare-access", "zero-trust", "cloudflared", "oidc", "skill-md", "security-ops" ], "source_url": "https://github.com/cloudflare/skills", "repo_url": "https://github.com/cloudflare/skills", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-cloudflare-one", "skill_markdown": "---\nname: cloudflare-one\ndescription: Design and audit Cloudflare Zero Trust Access applications, WARP device posture checks, identity provider OIDC claims, and Cloudflare Tunnel (`cloudflared`) routing. Use when working with Cloudflare Access, Zero Trust, cloudflared, OIDC.\ntriggers:\n - cloudflare one\n - cloudflare access\n - zero trust\n---\n\n# Cloudflare One Zero Trust & Access Policy Architecture Skill\n\n## 1. When to Use This Skill\nLoad `skill-cloudflare-one` whenever an engineering or operations task involves **Cloudflare Access, Zero Trust, cloudflared, OIDC**. Specifically:\n- Design and audit Cloudflare Zero Trust Access applications, WARP device posture checks, identity provider OIDC claims, and Cloudflare Tunnel (`cloudflared`) routing.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Cloudflare Access, Zero Trust, cloudflared, OIDC\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/cloudflare/skills`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Cloudflare Access**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\ncloudflared tunnel list\ncloudflared tunnel route dns app.example.com\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Cloudflare One Zero Trust & Access Policy Architecture Skill (skill-cloudflare-one) playbook. First inspect our current configuration and codebase for Cloudflare Access, Zero Trust, cloudflared, OIDC. Then propose a concrete, step-by-step implementation plan following the skill-cloudflare-one guardrails, run the non-destructive verification commands (cloudflared tunnel list), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Cloudflare Access", "Zero Trust", "cloudflared" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "cloudflare", "creator_name": "Cloudflare Developers", "creator_url": "https://x.com/cloudflare", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 86, "installs_count": 388, "agent_pulls_count": 834, "gravity_score": 3063.0, "is_featured": true, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_f8cb5101207f", "slug": "skill-cloudflare-one-migrations", "asset_type": "skill", "name": "Enterprise VPN to Cloudflare One Zero Trust Migration Playbook", "tagline": "Step-by-step migration blueprint from legacy VPN/SWG appliances to Cloudflare One: policy mapping, split-tunnel CIDR validation, and staged user cutover.", "description_md": "### Overview & Capability Scope\n**Enterprise VPN to Cloudflare One Zero Trust Migration Playbook** (`skill-cloudflare-one-migrations`) is a verified, portable `SKILL.md` playbook curated with **@cloudflare** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nStep-by-step migration blueprint from legacy VPN/SWG appliances to Cloudflare One: policy mapping, split-tunnel CIDR validation, and staged user cutover.\n\n### Required Stack & Tooling\n- **Primary Stack:** Zero Trust Migration, SWG, ZTNA, Cloudflare One\n- **Category:** `security-ops`\n- **Source & Reference:** [https://github.com/cloudflare/skills](https://github.com/cloudflare/skills)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-cloudflare-one-migrations\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-cloudflare-one-migrations/SKILL.md` or `~/.claude/skills/skill-cloudflare-one-migrations/SKILL.md`.\n\n### Verification Commands\n```bash\ncloudflared tunnel info\nnpx wrangler whoami\n```\n", "category": "security-ops", "subcategories": [ "security-ops", "skills" ], "tags": [ "zero-trust-migration", "swg", "ztna", "cloudflare-one", "skill-md", "security-ops" ], "source_url": "https://github.com/cloudflare/skills", "repo_url": "https://github.com/cloudflare/skills", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-cloudflare-one-migrations", "skill_markdown": "---\nname: cloudflare-one-migrations\ndescription: Step-by-step migration blueprint from legacy VPN/SWG appliances to Cloudflare One: policy mapping, split-tunnel CIDR validation, and staged user cutover. Use when working with Zero Trust Migration, SWG, ZTNA, Cloudflare One.\ntriggers:\n - cloudflare one migrations\n - zero trust migration\n - swg\n---\n\n# Enterprise VPN to Cloudflare One Zero Trust Migration Playbook\n\n## 1. When to Use This Skill\nLoad `skill-cloudflare-one-migrations` whenever an engineering or operations task involves **Zero Trust Migration, SWG, ZTNA, Cloudflare One**. Specifically:\n- Step-by-step migration blueprint from legacy VPN/SWG appliances to Cloudflare One: policy mapping, split-tunnel CIDR validation, and staged user cutover.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Zero Trust Migration, SWG, ZTNA, Cloudflare One\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/cloudflare/skills`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Zero Trust Migration**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\ncloudflared tunnel info\nnpx wrangler whoami\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Enterprise VPN to Cloudflare One Zero Trust Migration Playbook (skill-cloudflare-one-migrations) playbook. First inspect our current configuration and codebase for Zero Trust Migration, SWG, ZTNA, Cloudflare One. Then propose a concrete, step-by-step implementation plan following the skill-cloudflare-one-migrations guardrails, run the non-destructive verification commands (cloudflared tunnel info), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Zero Trust Migration", "SWG", "ZTNA" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "cloudflare", "creator_name": "Cloudflare Developers", "creator_url": "https://x.com/cloudflare", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 85, "installs_count": 384, "agent_pulls_count": 827, "gravity_score": 3034.0, "is_featured": true, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_fcdd7c5ab144", "slug": "skill-basin", "asset_type": "skill", "name": "Cloudflare Basin R2 Data Catalog & Iceberg Streaming Skill", "tagline": "Stream high-volume event telemetry into Apache Iceberg tables on Cloudflare R2 using Basin Pipelines, Basin Catalog, and distributed SQL query engines.", "description_md": "### Overview & Capability Scope\n**Cloudflare Basin R2 Data Catalog & Iceberg Streaming Skill** (`skill-basin`) is a verified, portable `SKILL.md` playbook curated with **@cloudflare** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nStream high-volume event telemetry into Apache Iceberg tables on Cloudflare R2 using Basin Pipelines, Basin Catalog, and distributed SQL query engines.\n\n### Required Stack & Tooling\n- **Primary Stack:** Cloudflare R2, Apache Iceberg, Basin SQL, Data Pipelines\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/cloudflare/skills](https://github.com/cloudflare/skills)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-basin\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-basin/SKILL.md` or `~/.claude/skills/skill-basin/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx wrangler pipelines list\nnpx wrangler r2 bucket list\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "cloudflare-r2", "apache-iceberg", "basin-sql", "data-pipelines", "skill-md", "engineering" ], "source_url": "https://github.com/cloudflare/skills", "repo_url": "https://github.com/cloudflare/skills", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-basin", "skill_markdown": "---\nname: basin\ndescription: Stream high-volume event telemetry into Apache Iceberg tables on Cloudflare R2 using Basin Pipelines, Basin Catalog, and distributed SQL query engines. Use when working with Cloudflare R2, Apache Iceberg, Basin SQL, Data Pipelines.\ntriggers:\n - basin\n - cloudflare r2\n - apache iceberg\n---\n\n# Cloudflare Basin R2 Data Catalog & Iceberg Streaming Skill\n\n## 1. When to Use This Skill\nLoad `skill-basin` whenever an engineering or operations task involves **Cloudflare R2, Apache Iceberg, Basin SQL, Data Pipelines**. Specifically:\n- Stream high-volume event telemetry into Apache Iceberg tables on Cloudflare R2 using Basin Pipelines, Basin Catalog, and distributed SQL query engines.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Cloudflare R2, Apache Iceberg, Basin SQL, Data Pipelines\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/cloudflare/skills`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Cloudflare R2**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx wrangler pipelines list\nnpx wrangler r2 bucket list\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Cloudflare Basin R2 Data Catalog & Iceberg Streaming Skill (skill-basin) playbook. First inspect our current configuration and codebase for Cloudflare R2, Apache Iceberg, Basin SQL, Data Pipelines. Then propose a concrete, step-by-step implementation plan following the skill-basin guardrails, run the non-destructive verification commands (npx wrangler pipelines list), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Cloudflare R2", "Apache Iceberg", "Basin SQL" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "cloudflare", "creator_name": "Cloudflare Developers", "creator_url": "https://x.com/cloudflare", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 84, "installs_count": 380, "agent_pulls_count": 820, "gravity_score": 3005.0, "is_featured": true, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_2856814687fd", "slug": "skill-k2", "asset_type": "skill", "name": "Cloudflare K2 Durable Log Streams & Event Subscription Skill", "tagline": "Configure high-throughput durable event logs with Cloudflare K2 Streams: producer batching, retention policies, consumer offsets, and Worker subscriptions.", "description_md": "### Overview & Capability Scope\n**Cloudflare K2 Durable Log Streams & Event Subscription Skill** (`skill-k2`) is a verified, portable `SKILL.md` playbook curated with **@cloudflare** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nConfigure high-throughput durable event logs with Cloudflare K2 Streams: producer batching, retention policies, consumer offsets, and Worker subscriptions.\n\n### Required Stack & Tooling\n- **Primary Stack:** Cloudflare K2, Event Streaming, Durable Logs, Workers\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/cloudflare/skills](https://github.com/cloudflare/skills)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-k2\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-k2/SKILL.md` or `~/.claude/skills/skill-k2/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx wrangler queues list\nnpx wrangler deploy\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "cloudflare-k2", "event-streaming", "durable-logs", "workers", "skill-md", "engineering" ], "source_url": "https://github.com/cloudflare/skills", "repo_url": "https://github.com/cloudflare/skills", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-k2", "skill_markdown": "---\nname: k2\ndescription: Configure high-throughput durable event logs with Cloudflare K2 Streams: producer batching, retention policies, consumer offsets, and Worker subscriptions. Use when working with Cloudflare K2, Event Streaming, Durable Logs, Workers.\ntriggers:\n - k2\n - cloudflare k2\n - event streaming\n---\n\n# Cloudflare K2 Durable Log Streams & Event Subscription Skill\n\n## 1. When to Use This Skill\nLoad `skill-k2` whenever an engineering or operations task involves **Cloudflare K2, Event Streaming, Durable Logs, Workers**. Specifically:\n- Configure high-throughput durable event logs with Cloudflare K2 Streams: producer batching, retention policies, consumer offsets, and Worker subscriptions.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Cloudflare K2, Event Streaming, Durable Logs, Workers\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/cloudflare/skills`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Cloudflare K2**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx wrangler queues list\nnpx wrangler deploy\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Cloudflare K2 Durable Log Streams & Event Subscription Skill (skill-k2) playbook. First inspect our current configuration and codebase for Cloudflare K2, Event Streaming, Durable Logs, Workers. Then propose a concrete, step-by-step implementation plan following the skill-k2 guardrails, run the non-destructive verification commands (npx wrangler queues list), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Cloudflare K2", "Event Streaming", "Durable Logs" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "cloudflare", "creator_name": "Cloudflare Developers", "creator_url": "https://x.com/cloudflare", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 83, "installs_count": 376, "agent_pulls_count": 813, "gravity_score": 2976.0, "is_featured": true, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_78a57c698bf5", "slug": "skill-stripe-best-practices", "asset_type": "skill", "name": "Stripe Checkout Sessions & PaymentIntents Architecture Skill", "tagline": "Official Stripe integration playbook covering Checkout Sessions vs PaymentIntents, dynamic payment methods, customer portal configuration, and PCI-complian", "description_md": "### Overview & Capability Scope\n**Stripe Checkout Sessions & PaymentIntents Architecture Skill** (`skill-stripe-best-practices`) is a verified, portable `SKILL.md` playbook curated with **@stripe** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nOfficial Stripe integration playbook covering Checkout Sessions vs PaymentIntents, dynamic payment methods, customer portal configuration, and PCI-compliant flows.\n\n### Required Stack & Tooling\n- **Primary Stack:** Stripe API, Checkout Sessions, PaymentIntents, Billing\n- **Category:** `finance-deals`\n- **Source & Reference:** [https://github.com/stripe/agent-toolkit](https://github.com/stripe/agent-toolkit)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-stripe-best-practices\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-stripe-best-practices/SKILL.md` or `~/.claude/skills/skill-stripe-best-practices/SKILL.md`.\n\n### Verification Commands\n```bash\nstripe listen --forward-to localhost:4242/webhook\nstripe trigger checkout.session.completed\n```\n", "category": "finance-deals", "subcategories": [ "finance-deals", "skills" ], "tags": [ "stripe-api", "checkout-sessions", "paymentintents", "billing", "skill-md", "finance-deals" ], "source_url": "https://github.com/stripe/agent-toolkit", "repo_url": "https://github.com/stripe/agent-toolkit", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-stripe-best-practices", "skill_markdown": "---\nname: stripe-best-practices\ndescription: Official Stripe integration playbook covering Checkout Sessions vs PaymentIntents, dynamic payment methods, customer portal configuration, and PCI-compliant flows. Use when working with Stripe API, Checkout Sessions, PaymentIntents, Billing.\ntriggers:\n - stripe best practices\n - stripe api\n - checkout sessions\n---\n\n# Stripe Checkout Sessions & PaymentIntents Architecture Skill\n\n## 1. When to Use This Skill\nLoad `skill-stripe-best-practices` whenever an engineering or operations task involves **Stripe API, Checkout Sessions, PaymentIntents, Billing**. Specifically:\n- Official Stripe integration playbook covering Checkout Sessions vs PaymentIntents, dynamic payment methods, customer portal configuration, and PCI-compliant flows.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Stripe API, Checkout Sessions, PaymentIntents, Billing\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/stripe/agent-toolkit`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Stripe API**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nstripe listen --forward-to localhost:4242/webhook\nstripe trigger checkout.session.completed\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Stripe Checkout Sessions & PaymentIntents Architecture Skill (skill-stripe-best-practices) playbook. First inspect our current configuration and codebase for Stripe API, Checkout Sessions, PaymentIntents, Billing. Then propose a concrete, step-by-step implementation plan following the skill-stripe-best-practices guardrails, run the non-destructive verification commands (stripe listen --forward-to localhost:4242/webhook), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Stripe API", "Checkout Sessions", "PaymentIntents" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "stripe", "creator_name": "Stripe Developer Team", "creator_url": "https://x.com/stripe", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 94, "installs_count": 372, "agent_pulls_count": 806, "gravity_score": 2932.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_3994bb83a0ad", "slug": "skill-stripe-webhooks", "asset_type": "skill", "name": "Stripe Webhook Signature Verification & Idempotency Playbook", "tagline": "Harden Stripe webhook endpoints with raw-body `Stripe-Signature` HMAC verification, idempotent event deduplication tables, and out-of-order event handling.", "description_md": "### Overview & Capability Scope\n**Stripe Webhook Signature Verification & Idempotency Playbook** (`skill-stripe-webhooks`) is a verified, portable `SKILL.md` playbook curated with **@stripe** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nHarden Stripe webhook endpoints with raw-body `Stripe-Signature` HMAC verification, idempotent event deduplication tables, and out-of-order event handling.\n\n### Required Stack & Tooling\n- **Primary Stack:** Stripe Webhooks, HMAC Verification, Idempotency, Postgres/D1\n- **Category:** `finance-deals`\n- **Source & Reference:** [https://github.com/stripe/agent-toolkit](https://github.com/stripe/agent-toolkit)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-stripe-webhooks\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-stripe-webhooks/SKILL.md` or `~/.claude/skills/skill-stripe-webhooks/SKILL.md`.\n\n### Verification Commands\n```bash\nstripe webhooks list\nstripe trigger payment_intent.succeeded\n```\n", "category": "finance-deals", "subcategories": [ "finance-deals", "skills" ], "tags": [ "stripe-webhooks", "hmac-verification", "idempotency", "postgres/d1", "skill-md", "finance-deals" ], "source_url": "https://github.com/stripe/agent-toolkit", "repo_url": "https://github.com/stripe/agent-toolkit", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-stripe-webhooks", "skill_markdown": "---\nname: stripe-webhooks\ndescription: Harden Stripe webhook endpoints with raw-body `Stripe-Signature` HMAC verification, idempotent event deduplication tables, and out-of-order event handling. Use when working with Stripe Webhooks, HMAC Verification, Idempotency, Postgres/D1.\ntriggers:\n - stripe webhooks\n - stripe webhooks\n - hmac verification\n---\n\n# Stripe Webhook Signature Verification & Idempotency Playbook\n\n## 1. When to Use This Skill\nLoad `skill-stripe-webhooks` whenever an engineering or operations task involves **Stripe Webhooks, HMAC Verification, Idempotency, Postgres/D1**. Specifically:\n- Harden Stripe webhook endpoints with raw-body `Stripe-Signature` HMAC verification, idempotent event deduplication tables, and out-of-order event handling.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Stripe Webhooks, HMAC Verification, Idempotency, Postgres/D1\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/stripe/agent-toolkit`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Stripe Webhooks**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nstripe webhooks list\nstripe trigger payment_intent.succeeded\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Stripe Webhook Signature Verification & Idempotency Playbook (skill-stripe-webhooks) playbook. First inspect our current configuration and codebase for Stripe Webhooks, HMAC Verification, Idempotency, Postgres/D1. Then propose a concrete, step-by-step implementation plan following the skill-stripe-webhooks guardrails, run the non-destructive verification commands (stripe webhooks list), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Stripe Webhooks", "HMAC Verification", "Idempotency" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "stripe", "creator_name": "Stripe Developer Team", "creator_url": "https://x.com/stripe", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 93, "installs_count": 368, "agent_pulls_count": 799, "gravity_score": 2903.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_d0076b7309aa", "slug": "skill-stripe-connect", "asset_type": "skill", "name": "Stripe Connect Multi-Party Marketplace & Payouts Playbook", "tagline": "Architect multi-vendor platforms with Stripe Connect: controller properties, destination vs direct charges, application fees, and automated KYC onboarding ", "description_md": "### Overview & Capability Scope\n**Stripe Connect Multi-Party Marketplace & Payouts Playbook** (`skill-stripe-connect`) is a verified, portable `SKILL.md` playbook curated with **@stripe** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nArchitect multi-vendor platforms with Stripe Connect: controller properties, destination vs direct charges, application fees, and automated KYC onboarding links.\n\n### Required Stack & Tooling\n- **Primary Stack:** Stripe Connect, Marketplace Payouts, KYC Onboarding, Split Payments\n- **Category:** `finance-deals`\n- **Source & Reference:** [https://github.com/stripe/agent-toolkit](https://github.com/stripe/agent-toolkit)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-stripe-connect\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-stripe-connect/SKILL.md` or `~/.claude/skills/skill-stripe-connect/SKILL.md`.\n\n### Verification Commands\n```bash\nstripe accounts create --type=express\nstripe transfers list\n```\n", "category": "finance-deals", "subcategories": [ "finance-deals", "skills" ], "tags": [ "stripe-connect", "marketplace-payouts", "kyc-onboarding", "split-payments", "skill-md", "finance-deals" ], "source_url": "https://github.com/stripe/agent-toolkit", "repo_url": "https://github.com/stripe/agent-toolkit", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-stripe-connect", "skill_markdown": "---\nname: stripe-connect\ndescription: Architect multi-vendor platforms with Stripe Connect: controller properties, destination vs direct charges, application fees, and automated KYC onboarding links. Use when working with Stripe Connect, Marketplace Payouts, KYC Onboarding, Split Payments.\ntriggers:\n - stripe connect\n - stripe connect\n - marketplace payouts\n---\n\n# Stripe Connect Multi-Party Marketplace & Payouts Playbook\n\n## 1. When to Use This Skill\nLoad `skill-stripe-connect` whenever an engineering or operations task involves **Stripe Connect, Marketplace Payouts, KYC Onboarding, Split Payments**. Specifically:\n- Architect multi-vendor platforms with Stripe Connect: controller properties, destination vs direct charges, application fees, and automated KYC onboarding links.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Stripe Connect, Marketplace Payouts, KYC Onboarding, Split Payments\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/stripe/agent-toolkit`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Stripe Connect**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nstripe accounts create --type=express\nstripe transfers list\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Stripe Connect Multi-Party Marketplace & Payouts Playbook (skill-stripe-connect) playbook. First inspect our current configuration and codebase for Stripe Connect, Marketplace Payouts, KYC Onboarding, Split Payments. Then propose a concrete, step-by-step implementation plan following the skill-stripe-connect guardrails, run the non-destructive verification commands (stripe accounts create --type=express), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Stripe Connect", "Marketplace Payouts", "KYC Onboarding" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "stripe", "creator_name": "Stripe Developer Team", "creator_url": "https://x.com/stripe", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 92, "installs_count": 364, "agent_pulls_count": 792, "gravity_score": 2874.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_604a1f86121c", "slug": "skill-stripe-subscriptions", "asset_type": "skill", "name": "Stripe SaaS Subscription Lifecycle, Trials & Proration Skill", "tagline": "Manage SaaS recurring billing: multi-tier price tables, trial-to-paid conversion webhooks, mid-cycle seat proration, and entitlement provisioning.", "description_md": "### Overview & Capability Scope\n**Stripe SaaS Subscription Lifecycle, Trials & Proration Skill** (`skill-stripe-subscriptions`) is a verified, portable `SKILL.md` playbook curated with **@stripe** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nManage SaaS recurring billing: multi-tier price tables, trial-to-paid conversion webhooks, mid-cycle seat proration, and entitlement provisioning.\n\n### Required Stack & Tooling\n- **Primary Stack:** Stripe Billing, Subscriptions, Entitlements, SaaS Metering\n- **Category:** `finance-deals`\n- **Source & Reference:** [https://github.com/stripe/agent-toolkit](https://github.com/stripe/agent-toolkit)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-stripe-subscriptions\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-stripe-subscriptions/SKILL.md` or `~/.claude/skills/skill-stripe-subscriptions/SKILL.md`.\n\n### Verification Commands\n```bash\nstripe subscriptions list --limit=10\nstripe invoices upcoming --customer=cus_123\n```\n", "category": "finance-deals", "subcategories": [ "finance-deals", "skills" ], "tags": [ "stripe-billing", "subscriptions", "entitlements", "saas-metering", "skill-md", "finance-deals" ], "source_url": "https://github.com/stripe/agent-toolkit", "repo_url": "https://github.com/stripe/agent-toolkit", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-stripe-subscriptions", "skill_markdown": "---\nname: stripe-subscriptions\ndescription: Manage SaaS recurring billing: multi-tier price tables, trial-to-paid conversion webhooks, mid-cycle seat proration, and entitlement provisioning. Use when working with Stripe Billing, Subscriptions, Entitlements, SaaS Metering.\ntriggers:\n - stripe subscriptions\n - stripe billing\n - subscriptions\n---\n\n# Stripe SaaS Subscription Lifecycle, Trials & Proration Skill\n\n## 1. When to Use This Skill\nLoad `skill-stripe-subscriptions` whenever an engineering or operations task involves **Stripe Billing, Subscriptions, Entitlements, SaaS Metering**. Specifically:\n- Manage SaaS recurring billing: multi-tier price tables, trial-to-paid conversion webhooks, mid-cycle seat proration, and entitlement provisioning.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Stripe Billing, Subscriptions, Entitlements, SaaS Metering\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/stripe/agent-toolkit`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Stripe Billing**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nstripe subscriptions list --limit=10\nstripe invoices upcoming --customer=cus_123\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Stripe SaaS Subscription Lifecycle, Trials & Proration Skill (skill-stripe-subscriptions) playbook. First inspect our current configuration and codebase for Stripe Billing, Subscriptions, Entitlements, SaaS Metering. Then propose a concrete, step-by-step implementation plan following the skill-stripe-subscriptions guardrails, run the non-destructive verification commands (stripe subscriptions list --limit=10), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Stripe Billing", "Subscriptions", "Entitlements" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "stripe", "creator_name": "Stripe Developer Team", "creator_url": "https://x.com/stripe", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 91, "installs_count": 360, "agent_pulls_count": 785, "gravity_score": 2845.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_98bb3c34cfce", "slug": "skill-metronome-billing", "asset_type": "skill", "name": "Metronome Usage-Based Metering & Credit Commit Billing Skill", "tagline": "Design usage-based AI & infrastructure billing with Metronome: idempotent usage event ingestion, SQL/dimensional billable metrics, rate cards, and prepaid ", "description_md": "### Overview & Capability Scope\n**Metronome Usage-Based Metering & Credit Commit Billing Skill** (`skill-metronome-billing`) is a verified, portable `SKILL.md` playbook curated with **@stripe** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nDesign usage-based AI & infrastructure billing with Metronome: idempotent usage event ingestion, SQL/dimensional billable metrics, rate cards, and prepaid commits.\n\n### Required Stack & Tooling\n- **Primary Stack:** Metronome API, Usage-Based Billing, Idempotency Keys, Rate Cards\n- **Category:** `finance-deals`\n- **Source & Reference:** [https://docs.metronome.com](https://docs.metronome.com)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-metronome-billing\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-metronome-billing/SKILL.md` or `~/.claude/skills/skill-metronome-billing/SKILL.md`.\n\n### Verification Commands\n```bash\ncurl -X POST https://api.metronome.com/v1/ingest -H 'Authorization: Bearer $METRONOME_KEY'\n```\n", "category": "finance-deals", "subcategories": [ "finance-deals", "skills" ], "tags": [ "metronome-api", "usage-based-billing", "idempotency-keys", "rate-cards", "skill-md", "finance-deals" ], "source_url": "https://docs.metronome.com", "repo_url": "https://docs.metronome.com", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-metronome-billing", "skill_markdown": "---\nname: metronome-billing\ndescription: Design usage-based AI & infrastructure billing with Metronome: idempotent usage event ingestion, SQL/dimensional billable metrics, rate cards, and prepaid commits. Use when working with Metronome API, Usage-Based Billing, Idempotency Keys, Rate Cards.\ntriggers:\n - metronome billing\n - metronome api\n - usage-based billing\n---\n\n# Metronome Usage-Based Metering & Credit Commit Billing Skill\n\n## 1. When to Use This Skill\nLoad `skill-metronome-billing` whenever an engineering or operations task involves **Metronome API, Usage-Based Billing, Idempotency Keys, Rate Cards**. Specifically:\n- Design usage-based AI & infrastructure billing with Metronome: idempotent usage event ingestion, SQL/dimensional billable metrics, rate cards, and prepaid commits.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Metronome API, Usage-Based Billing, Idempotency Keys, Rate Cards\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://docs.metronome.com`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Metronome API**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\ncurl -X POST https://api.metronome.com/v1/ingest -H 'Authorization: Bearer $METRONOME_KEY'\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Metronome Usage-Based Metering & Credit Commit Billing Skill (skill-metronome-billing) playbook. First inspect our current configuration and codebase for Metronome API, Usage-Based Billing, Idempotency Keys, Rate Cards. Then propose a concrete, step-by-step implementation plan following the skill-metronome-billing guardrails, run the non-destructive verification commands (curl -X POST https://api.metronome.com/v1/ingest -H 'Authorization: Bearer $METRONOME_KEY'), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Metronome API", "Usage-Based Billing", "Idempotency Keys" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "stripe", "creator_name": "Stripe Developer Team", "creator_url": "https://x.com/stripe", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 90, "installs_count": 356, "agent_pulls_count": 778, "gravity_score": 2816.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_396364496c9c", "slug": "skill-stripe-dunning-recovery", "asset_type": "skill", "name": "Autonomous SaaS Failed-Payment Dunning & Churn Recovery Skill", "tagline": "Detect `invoice.payment_failed` events, classify decline codes (soft vs hard declines), generate hosted card-update links, and stage personalized recovery ", "description_md": "### Overview & Capability Scope\n**Autonomous SaaS Failed-Payment Dunning & Churn Recovery Skill** (`skill-stripe-dunning-recovery`) is a verified, portable `SKILL.md` playbook curated with **@gergelyorosz** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nDetect `invoice.payment_failed` events, classify decline codes (soft vs hard declines), generate hosted card-update links, and stage personalized recovery outreach.\n\n### Required Stack & Tooling\n- **Primary Stack:** Stripe Billing, Churn Recovery, Payment Links, Gmail/Slack\n- **Category:** `finance-deals`\n- **Source & Reference:** [https://github.com/stripe/agent-toolkit](https://github.com/stripe/agent-toolkit)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-stripe-dunning-recovery\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-stripe-dunning-recovery/SKILL.md` or `~/.claude/skills/skill-stripe-dunning-recovery/SKILL.md`.\n\n### Verification Commands\n```bash\nstripe events list --type=invoice.payment_failed\nstripe billing_portal sessions create\n```\n", "category": "finance-deals", "subcategories": [ "finance-deals", "skills" ], "tags": [ "stripe-billing", "churn-recovery", "payment-links", "gmail/slack", "skill-md", "finance-deals" ], "source_url": "https://github.com/stripe/agent-toolkit", "repo_url": "https://github.com/stripe/agent-toolkit", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-stripe-dunning-recovery", "skill_markdown": "---\nname: stripe-dunning-recovery\ndescription: Detect `invoice.payment_failed` events, classify decline codes (soft vs hard declines), generate hosted card-update links, and stage personalized recovery outreach. Use when working with Stripe Billing, Churn Recovery, Payment Links, Gmail/Slack.\ntriggers:\n - stripe dunning recovery\n - stripe billing\n - churn recovery\n---\n\n# Autonomous SaaS Failed-Payment Dunning & Churn Recovery Skill\n\n## 1. When to Use This Skill\nLoad `skill-stripe-dunning-recovery` whenever an engineering or operations task involves **Stripe Billing, Churn Recovery, Payment Links, Gmail/Slack**. Specifically:\n- Detect `invoice.payment_failed` events, classify decline codes (soft vs hard declines), generate hosted card-update links, and stage personalized recovery outreach.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Stripe Billing, Churn Recovery, Payment Links, Gmail/Slack\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/stripe/agent-toolkit`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Stripe Billing**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nstripe events list --type=invoice.payment_failed\nstripe billing_portal sessions create\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Autonomous SaaS Failed-Payment Dunning & Churn Recovery Skill (skill-stripe-dunning-recovery) playbook. First inspect our current configuration and codebase for Stripe Billing, Churn Recovery, Payment Links, Gmail/Slack. Then propose a concrete, step-by-step implementation plan following the skill-stripe-dunning-recovery guardrails, run the non-destructive verification commands (stripe events list --type=invoice.payment_failed), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Stripe Billing", "Churn Recovery", "Payment Links" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "gergelyorosz", "creator_name": "Gergely Orosz (@GergelyOrosz)", "creator_url": "https://x.com/gergelyorosz", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 89, "installs_count": 352, "agent_pulls_count": 771, "gravity_score": 2787.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_83f9806f0304", "slug": "skill-upgrade-stripe", "asset_type": "skill", "name": "Stripe API Version & SDK Zero-Regression Upgrade Playbook", "tagline": "Safely upgrade Stripe API versions (`2026-xx-xx`) and server SDKs: inspect breaking changelog diffs, pin webhook endpoint versions, and run sandbox replay ", "description_md": "### Overview & Capability Scope\n**Stripe API Version & SDK Zero-Regression Upgrade Playbook** (`skill-upgrade-stripe`) is a verified, portable `SKILL.md` playbook curated with **@stripe** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nSafely upgrade Stripe API versions (`2026-xx-xx`) and server SDKs: inspect breaking changelog diffs, pin webhook endpoint versions, and run sandbox replay tests.\n\n### Required Stack & Tooling\n- **Primary Stack:** Stripe SDK, API Versioning, Webhook Testing, TypeScript\n- **Category:** `engineering`\n- **Source & Reference:** [https://docs.stripe.com/upgrades](https://docs.stripe.com/upgrades)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-upgrade-stripe\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-upgrade-stripe/SKILL.md` or `~/.claude/skills/skill-upgrade-stripe/SKILL.md`.\n\n### Verification Commands\n```bash\nnpm i stripe@latest\nstripe listen --latest\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "stripe-sdk", "api-versioning", "webhook-testing", "typescript", "skill-md", "engineering" ], "source_url": "https://docs.stripe.com/upgrades", "repo_url": "https://docs.stripe.com/upgrades", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-upgrade-stripe", "skill_markdown": "---\nname: upgrade-stripe\ndescription: Safely upgrade Stripe API versions (`2026-xx-xx`) and server SDKs: inspect breaking changelog diffs, pin webhook endpoint versions, and run sandbox replay tests. Use when working with Stripe SDK, API Versioning, Webhook Testing, TypeScript.\ntriggers:\n - upgrade stripe\n - stripe sdk\n - api versioning\n---\n\n# Stripe API Version & SDK Zero-Regression Upgrade Playbook\n\n## 1. When to Use This Skill\nLoad `skill-upgrade-stripe` whenever an engineering or operations task involves **Stripe SDK, API Versioning, Webhook Testing, TypeScript**. Specifically:\n- Safely upgrade Stripe API versions (`2026-xx-xx`) and server SDKs: inspect breaking changelog diffs, pin webhook endpoint versions, and run sandbox replay tests.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Stripe SDK, API Versioning, Webhook Testing, TypeScript\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://docs.stripe.com/upgrades`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Stripe SDK**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpm i stripe@latest\nstripe listen --latest\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Stripe API Version & SDK Zero-Regression Upgrade Playbook (skill-upgrade-stripe) playbook. First inspect our current configuration and codebase for Stripe SDK, API Versioning, Webhook Testing, TypeScript. Then propose a concrete, step-by-step implementation plan following the skill-upgrade-stripe guardrails, run the non-destructive verification commands (npm i stripe@latest), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Stripe SDK", "API Versioning", "Webhook Testing" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "stripe", "creator_name": "Stripe Developer Team", "creator_url": "https://x.com/stripe", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 88, "installs_count": 348, "agent_pulls_count": 764, "gravity_score": 2758.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_9c90038513f8", "slug": "skill-supabase-rls-audit", "asset_type": "skill", "name": "Supabase Postgres Row-Level Security (RLS) Audit Playbook", "tagline": "Audit every table in a Supabase `public` schema for missing RLS policies, insecure `USING (true)` clauses, unindexed foreign keys, and leaked `service_role", "description_md": "### Overview & Capability Scope\n**Supabase Postgres Row-Level Security (RLS) Audit Playbook** (`skill-supabase-rls-audit`) is a verified, portable `SKILL.md` playbook curated with **@supabase** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nAudit every table in a Supabase `public` schema for missing RLS policies, insecure `USING (true)` clauses, unindexed foreign keys, and leaked `service_role` keys.\n\n### Required Stack & Tooling\n- **Primary Stack:** Supabase, PostgreSQL, Row-Level Security, SQL Audit\n- **Category:** `security-ops`\n- **Source & Reference:** [https://github.com/supabase-community/supabase-mcp](https://github.com/supabase-community/supabase-mcp)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-supabase-rls-audit\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-supabase-rls-audit/SKILL.md` or `~/.claude/skills/skill-supabase-rls-audit/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx supabase db lint\nnpx supabase inspect db table-stats\n```\n", "category": "security-ops", "subcategories": [ "security-ops", "skills" ], "tags": [ "supabase", "postgresql", "row-level-security", "sql-audit", "skill-md", "security-ops" ], "source_url": "https://github.com/supabase-community/supabase-mcp", "repo_url": "https://github.com/supabase-community/supabase-mcp", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-supabase-rls-audit", "skill_markdown": "---\nname: supabase-rls-audit\ndescription: Audit every table in a Supabase `public` schema for missing RLS policies, insecure `USING (true)` clauses, unindexed foreign keys, and leaked `service_role` keys. Use when working with Supabase, PostgreSQL, Row-Level Security, SQL Audit.\ntriggers:\n - supabase rls audit\n - supabase\n - postgresql\n---\n\n# Supabase Postgres Row-Level Security (RLS) Audit Playbook\n\n## 1. When to Use This Skill\nLoad `skill-supabase-rls-audit` whenever an engineering or operations task involves **Supabase, PostgreSQL, Row-Level Security, SQL Audit**. Specifically:\n- Audit every table in a Supabase `public` schema for missing RLS policies, insecure `USING (true)` clauses, unindexed foreign keys, and leaked `service_role` keys.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Supabase, PostgreSQL, Row-Level Security, SQL Audit\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/supabase-community/supabase-mcp`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Supabase**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx supabase db lint\nnpx supabase inspect db table-stats\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Supabase Postgres Row-Level Security (RLS) Audit Playbook (skill-supabase-rls-audit) playbook. First inspect our current configuration and codebase for Supabase, PostgreSQL, Row-Level Security, SQL Audit. Then propose a concrete, step-by-step implementation plan following the skill-supabase-rls-audit guardrails, run the non-destructive verification commands (npx supabase db lint), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Supabase", "PostgreSQL", "Row-Level Security" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "supabase", "creator_name": "Supabase Engineering", "creator_url": "https://x.com/supabase", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 87, "installs_count": 344, "agent_pulls_count": 757, "gravity_score": 2729.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_a57c2b8dbf46", "slug": "skill-zero-downtime-migrations", "asset_type": "skill", "name": "Prisma & Drizzle Zero-Downtime Expand/Contract Migration Skill", "tagline": "Execute expand-and-contract Postgres schema migrations without table locks: nullable column addition, batched background backfill, concurrent index creatio", "description_md": "### Overview & Capability Scope\n**Prisma & Drizzle Zero-Downtime Expand/Contract Migration Skill** (`skill-zero-downtime-migrations`) is a verified, portable `SKILL.md` playbook curated with **@mattpocockuk** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nExecute expand-and-contract Postgres schema migrations without table locks: nullable column addition, batched background backfill, concurrent index creation, and safe drop.\n\n### Required Stack & Tooling\n- **Primary Stack:** Drizzle ORM, Prisma, PostgreSQL, Zero-Downtime Migrations\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/drizzle-team/drizzle-orm](https://github.com/drizzle-team/drizzle-orm)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-zero-downtime-migrations\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-zero-downtime-migrations/SKILL.md` or `~/.claude/skills/skill-zero-downtime-migrations/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx drizzle-kit generate\nnpx drizzle-kit check\nnpx prisma migrate diff\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "drizzle-orm", "prisma", "postgresql", "zero-downtime-migrations", "skill-md", "engineering" ], "source_url": "https://github.com/drizzle-team/drizzle-orm", "repo_url": "https://github.com/drizzle-team/drizzle-orm", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-zero-downtime-migrations", "skill_markdown": "---\nname: zero-downtime-migrations\ndescription: Execute expand-and-contract Postgres schema migrations without table locks: nullable column addition, batched background backfill, concurrent index creation, and safe drop. Use when working with Drizzle ORM, Prisma, PostgreSQL, Zero-Downtime Migrations.\ntriggers:\n - zero downtime migrations\n - drizzle orm\n - prisma\n---\n\n# Prisma & Drizzle Zero-Downtime Expand/Contract Migration Skill\n\n## 1. When to Use This Skill\nLoad `skill-zero-downtime-migrations` whenever an engineering or operations task involves **Drizzle ORM, Prisma, PostgreSQL, Zero-Downtime Migrations**. Specifically:\n- Execute expand-and-contract Postgres schema migrations without table locks: nullable column addition, batched background backfill, concurrent index creation, and safe drop.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Drizzle ORM, Prisma, PostgreSQL, Zero-Downtime Migrations\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/drizzle-team/drizzle-orm`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Drizzle ORM**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx drizzle-kit generate\nnpx drizzle-kit check\nnpx prisma migrate diff\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Prisma & Drizzle Zero-Downtime Expand/Contract Migration Skill (skill-zero-downtime-migrations) playbook. First inspect our current configuration and codebase for Drizzle ORM, Prisma, PostgreSQL, Zero-Downtime Migrations. Then propose a concrete, step-by-step implementation plan following the skill-zero-downtime-migrations guardrails, run the non-destructive verification commands (npx drizzle-kit generate), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Drizzle ORM", "Prisma", "PostgreSQL" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "mattpocockuk", "creator_name": "Matt Pocock (@mattpocockuk)", "creator_url": "https://x.com/mattpocockuk", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 86, "installs_count": 340, "agent_pulls_count": 750, "gravity_score": 2700.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_5f71a81512ec", "slug": "skill-upstash-ratelimit", "asset_type": "skill", "name": "Upstash Redis Sliding-Window Rate Limiting & Edge Cache Skill", "tagline": "Protect public APIs and LLM endpoints from abuse with Upstash Redis sliding-window and token-bucket rate limiters at the Cloudflare Worker or Next.js edge.", "description_md": "### Overview & Capability Scope\n**Upstash Redis Sliding-Window Rate Limiting & Edge Cache Skill** (`skill-upstash-ratelimit`) is a verified, portable `SKILL.md` playbook curated with **@rauchg** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nProtect public APIs and LLM endpoints from abuse with Upstash Redis sliding-window and token-bucket rate limiters at the Cloudflare Worker or Next.js edge.\n\n### Required Stack & Tooling\n- **Primary Stack:** Upstash Redis, @upstash/ratelimit, Edge Middleware, API Protection\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/upstash/ratelimit-js](https://github.com/upstash/ratelimit-js)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-upstash-ratelimit\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-upstash-ratelimit/SKILL.md` or `~/.claude/skills/skill-upstash-ratelimit/SKILL.md`.\n\n### Verification Commands\n```bash\nnpm i @upstash/redis @upstash/ratelimit\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "upstash-redis", "@upstash/ratelimit", "edge-middleware", "api-protection", "skill-md", "engineering" ], "source_url": "https://github.com/upstash/ratelimit-js", "repo_url": "https://github.com/upstash/ratelimit-js", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-upstash-ratelimit", "skill_markdown": "---\nname: upstash-ratelimit\ndescription: Protect public APIs and LLM endpoints from abuse with Upstash Redis sliding-window and token-bucket rate limiters at the Cloudflare Worker or Next.js edge. Use when working with Upstash Redis, @upstash/ratelimit, Edge Middleware, API Protection.\ntriggers:\n - upstash ratelimit\n - upstash redis\n - @upstash/ratelimit\n---\n\n# Upstash Redis Sliding-Window Rate Limiting & Edge Cache Skill\n\n## 1. When to Use This Skill\nLoad `skill-upstash-ratelimit` whenever an engineering or operations task involves **Upstash Redis, @upstash/ratelimit, Edge Middleware, API Protection**. Specifically:\n- Protect public APIs and LLM endpoints from abuse with Upstash Redis sliding-window and token-bucket rate limiters at the Cloudflare Worker or Next.js edge.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Upstash Redis, @upstash/ratelimit, Edge Middleware, API Protection\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/upstash/ratelimit-js`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Upstash Redis**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpm i @upstash/redis @upstash/ratelimit\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Upstash Redis Sliding-Window Rate Limiting & Edge Cache Skill (skill-upstash-ratelimit) playbook. First inspect our current configuration and codebase for Upstash Redis, @upstash/ratelimit, Edge Middleware, API Protection. Then propose a concrete, step-by-step implementation plan following the skill-upstash-ratelimit guardrails, run the non-destructive verification commands (npm i @upstash/redis @upstash/ratelimit), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Upstash Redis", "@upstash/ratelimit", "Edge Middleware" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "rauchg", "creator_name": "Guillermo Rauch (@rauchg)", "creator_url": "https://x.com/rauchg", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 85, "installs_count": 336, "agent_pulls_count": 743, "gravity_score": 2671.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_2fc049467e64", "slug": "skill-postgres-query-tuning", "asset_type": "skill", "name": "PostgreSQL EXPLAIN ANALYZE & Composite Index Optimization Skill", "tagline": "Diagnose slow SQL queries using `EXPLAIN (ANALYZE, BUFFERS, FORMAT JSON)`, eliminate sequential scans, design covering B-Tree/GIN indexes, and tune connect", "description_md": "### Overview & Capability Scope\n**PostgreSQL EXPLAIN ANALYZE & Composite Index Optimization Skill** (`skill-postgres-query-tuning`) is a verified, portable `SKILL.md` playbook curated with **@simonw** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nDiagnose slow SQL queries using `EXPLAIN (ANALYZE, BUFFERS, FORMAT JSON)`, eliminate sequential scans, design covering B-Tree/GIN indexes, and tune connection pools.\n\n### Required Stack & Tooling\n- **Primary Stack:** PostgreSQL, EXPLAIN ANALYZE, Indexing, pg_stat_statements\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/modelcontextprotocol/servers](https://github.com/modelcontextprotocol/servers)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-postgres-query-tuning\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-postgres-query-tuning/SKILL.md` or `~/.claude/skills/skill-postgres-query-tuning/SKILL.md`.\n\n### Verification Commands\n```bash\npsql $DATABASE_URL -c 'EXPLAIN (ANALYZE, BUFFERS) SELECT ...'\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "postgresql", "explain-analyze", "indexing", "pg_stat_statements", "skill-md", "engineering" ], "source_url": "https://github.com/modelcontextprotocol/servers", "repo_url": "https://github.com/modelcontextprotocol/servers", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-postgres-query-tuning", "skill_markdown": "---\nname: postgres-query-tuning\ndescription: Diagnose slow SQL queries using `EXPLAIN (ANALYZE, BUFFERS, FORMAT JSON)`, eliminate sequential scans, design covering B-Tree/GIN indexes, and tune connection pools. Use when working with PostgreSQL, EXPLAIN ANALYZE, Indexing, pg_stat_statements.\ntriggers:\n - postgres query tuning\n - postgresql\n - explain analyze\n---\n\n# PostgreSQL EXPLAIN ANALYZE & Composite Index Optimization Skill\n\n## 1. When to Use This Skill\nLoad `skill-postgres-query-tuning` whenever an engineering or operations task involves **PostgreSQL, EXPLAIN ANALYZE, Indexing, pg_stat_statements**. Specifically:\n- Diagnose slow SQL queries using `EXPLAIN (ANALYZE, BUFFERS, FORMAT JSON)`, eliminate sequential scans, design covering B-Tree/GIN indexes, and tune connection pools.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** PostgreSQL, EXPLAIN ANALYZE, Indexing, pg_stat_statements\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/modelcontextprotocol/servers`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **PostgreSQL**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\npsql $DATABASE_URL -c 'EXPLAIN (ANALYZE, BUFFERS) SELECT ...'\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the PostgreSQL EXPLAIN ANALYZE & Composite Index Optimization Skill (skill-postgres-query-tuning) playbook. First inspect our current configuration and codebase for PostgreSQL, EXPLAIN ANALYZE, Indexing, pg_stat_statements. Then propose a concrete, step-by-step implementation plan following the skill-postgres-query-tuning guardrails, run the non-destructive verification commands (psql $DATABASE_URL -c 'EXPLAIN (ANALYZE, BUFFERS) SELECT ...'), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "PostgreSQL", "EXPLAIN ANALYZE", "Indexing" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "simonw", "creator_name": "Simon Willison (@simonw)", "creator_url": "https://x.com/simonw", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 84, "installs_count": 332, "agent_pulls_count": 736, "gravity_score": 2642.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_7da35a1cc901", "slug": "skill-typescript-zod-validation", "asset_type": "skill", "name": "Strict TypeScript AST & Runtime Zod Boundary Validation Skill", "tagline": "Enforce end-to-end type safety across API boundaries, environment variables, and LLM structured outputs using strict `tsconfig.json` flags and composable Z", "description_md": "### Overview & Capability Scope\n**Strict TypeScript AST & Runtime Zod Boundary Validation Skill** (`skill-typescript-zod-validation`) is a verified, portable `SKILL.md` playbook curated with **@mattpocockuk** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nEnforce end-to-end type safety across API boundaries, environment variables, and LLM structured outputs using strict `tsconfig.json` flags and composable Zod schemas.\n\n### Required Stack & Tooling\n- **Primary Stack:** TypeScript 5+, Zod, Structured Outputs, Strict Mode\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/colinhacks/zod](https://github.com/colinhacks/zod)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-typescript-zod-validation\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-typescript-zod-validation/SKILL.md` or `~/.claude/skills/skill-typescript-zod-validation/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx tsc --noEmit --strict\nnpm test\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "typescript-5+", "zod", "structured-outputs", "strict-mode", "skill-md", "engineering" ], "source_url": "https://github.com/colinhacks/zod", "repo_url": "https://github.com/colinhacks/zod", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-typescript-zod-validation", "skill_markdown": "---\nname: typescript-zod-validation\ndescription: Enforce end-to-end type safety across API boundaries, environment variables, and LLM structured outputs using strict `tsconfig.json` flags and composable Zod schemas. Use when working with TypeScript 5+, Zod, Structured Outputs, Strict Mode.\ntriggers:\n - typescript zod validation\n - typescript 5+\n - zod\n---\n\n# Strict TypeScript AST & Runtime Zod Boundary Validation Skill\n\n## 1. When to Use This Skill\nLoad `skill-typescript-zod-validation` whenever an engineering or operations task involves **TypeScript 5+, Zod, Structured Outputs, Strict Mode**. Specifically:\n- Enforce end-to-end type safety across API boundaries, environment variables, and LLM structured outputs using strict `tsconfig.json` flags and composable Zod schemas.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** TypeScript 5+, Zod, Structured Outputs, Strict Mode\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/colinhacks/zod`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **TypeScript 5+**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx tsc --noEmit --strict\nnpm test\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Strict TypeScript AST & Runtime Zod Boundary Validation Skill (skill-typescript-zod-validation) playbook. First inspect our current configuration and codebase for TypeScript 5+, Zod, Structured Outputs, Strict Mode. Then propose a concrete, step-by-step implementation plan following the skill-typescript-zod-validation guardrails, run the non-destructive verification commands (npx tsc --noEmit --strict), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "TypeScript 5+", "Zod", "Structured Outputs" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "mattpocockuk", "creator_name": "Matt Pocock (@mattpocockuk)", "creator_url": "https://x.com/mattpocockuk", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 94, "installs_count": 324, "agent_pulls_count": 722, "gravity_score": 2614.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_b9e47b38ebe3", "slug": "skill-graphql-n1-prevention", "asset_type": "skill", "name": "GraphQL Schema Validation, Complexity Limits & DataLoader Skill", "tagline": "Eliminate N+1 database queries in GraphQL resolvers using request-scoped DataLoaders, enforce query depth/cost limits, and validate persisted operations.", "description_md": "### Overview & Capability Scope\n**GraphQL Schema Validation, Complexity Limits & DataLoader Skill** (`skill-graphql-n1-prevention`) is a verified, portable `SKILL.md` playbook curated with **@kentcdodds** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nEliminate N+1 database queries in GraphQL resolvers using request-scoped DataLoaders, enforce query depth/cost limits, and validate persisted operations.\n\n### Required Stack & Tooling\n- **Primary Stack:** GraphQL, DataLoader, TypeScript, API Security\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/graphql/dataloader](https://github.com/graphql/dataloader)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-graphql-n1-prevention\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-graphql-n1-prevention/SKILL.md` or `~/.claude/skills/skill-graphql-n1-prevention/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx graphql-inspector validate schema.graphql operations.graphql\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "graphql", "dataloader", "typescript", "api-security", "skill-md", "engineering" ], "source_url": "https://github.com/graphql/dataloader", "repo_url": "https://github.com/graphql/dataloader", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-graphql-n1-prevention", "skill_markdown": "---\nname: graphql-n1-prevention\ndescription: Eliminate N+1 database queries in GraphQL resolvers using request-scoped DataLoaders, enforce query depth/cost limits, and validate persisted operations. Use when working with GraphQL, DataLoader, TypeScript, API Security.\ntriggers:\n - graphql n1 prevention\n - graphql\n - dataloader\n---\n\n# GraphQL Schema Validation, Complexity Limits & DataLoader Skill\n\n## 1. When to Use This Skill\nLoad `skill-graphql-n1-prevention` whenever an engineering or operations task involves **GraphQL, DataLoader, TypeScript, API Security**. Specifically:\n- Eliminate N+1 database queries in GraphQL resolvers using request-scoped DataLoaders, enforce query depth/cost limits, and validate persisted operations.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** GraphQL, DataLoader, TypeScript, API Security\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/graphql/dataloader`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **GraphQL**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx graphql-inspector validate schema.graphql operations.graphql\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the GraphQL Schema Validation, Complexity Limits & DataLoader Skill (skill-graphql-n1-prevention) playbook. First inspect our current configuration and codebase for GraphQL, DataLoader, TypeScript, API Security. Then propose a concrete, step-by-step implementation plan following the skill-graphql-n1-prevention guardrails, run the non-destructive verification commands (npx graphql-inspector validate schema.graphql operations.graphql), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "GraphQL", "DataLoader", "TypeScript" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "kentcdodds", "creator_name": "Kent C. Dodds (@kentcdodds)", "creator_url": "https://x.com/kentcdodds", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 83, "installs_count": 328, "agent_pulls_count": 729, "gravity_score": 2613.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_fe4d001590fd", "slug": "skill-prompt-injection-defense", "asset_type": "skill", "name": "LLM Indirect Prompt Injection & Untrusted Data Fencing Skill", "tagline": "Defend autonomous agents against indirect prompt injection in emails, web pages, and MCP tool outputs using XML data fencing, privilege separation, and hum", "description_md": "### Overview & Capability Scope\n**LLM Indirect Prompt Injection & Untrusted Data Fencing Skill** (`skill-prompt-injection-defense`) is a verified, portable `SKILL.md` playbook curated with **@simonw** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nDefend autonomous agents against indirect prompt injection in emails, web pages, and MCP tool outputs using XML data fencing, privilege separation, and human approval gates.\n\n### Required Stack & Tooling\n- **Primary Stack:** Prompt Security, Untrusted Data Fencing, Tool Governance, Agent Safety\n- **Category:** `security-ops`\n- **Source & Reference:** [https://simonwillison.net/tags/prompt-injection/](https://simonwillison.net/tags/prompt-injection/)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-prompt-injection-defense\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-prompt-injection-defense/SKILL.md` or `~/.claude/skills/skill-prompt-injection-defense/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx lightstate inspect inbox-injection-guard\n```\n", "category": "security-ops", "subcategories": [ "security-ops", "skills" ], "tags": [ "prompt-security", "untrusted-data-fencing", "tool-governance", "agent-safety", "skill-md", "security-ops" ], "source_url": "https://simonwillison.net/tags/prompt-injection/", "repo_url": "https://simonwillison.net/tags/prompt-injection/", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-prompt-injection-defense", "skill_markdown": "---\nname: prompt-injection-defense\ndescription: Defend autonomous agents against indirect prompt injection in emails, web pages, and MCP tool outputs using XML data fencing, privilege separation, and human approval gates. Use when working with Prompt Security, Untrusted Data Fencing, Tool Governance, Agent Safety.\ntriggers:\n - prompt injection defense\n - prompt security\n - untrusted data fencing\n---\n\n# LLM Indirect Prompt Injection & Untrusted Data Fencing Skill\n\n## 1. When to Use This Skill\nLoad `skill-prompt-injection-defense` whenever an engineering or operations task involves **Prompt Security, Untrusted Data Fencing, Tool Governance, Agent Safety**. Specifically:\n- Defend autonomous agents against indirect prompt injection in emails, web pages, and MCP tool outputs using XML data fencing, privilege separation, and human approval gates.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Prompt Security, Untrusted Data Fencing, Tool Governance, Agent Safety\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://simonwillison.net/tags/prompt-injection/`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Prompt Security**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx lightstate inspect inbox-injection-guard\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the LLM Indirect Prompt Injection & Untrusted Data Fencing Skill (skill-prompt-injection-defense) playbook. First inspect our current configuration and codebase for Prompt Security, Untrusted Data Fencing, Tool Governance, Agent Safety. Then propose a concrete, step-by-step implementation plan following the skill-prompt-injection-defense guardrails, run the non-destructive verification commands (npx lightstate inspect inbox-injection-guard), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Prompt Security", "Untrusted Data Fencing", "Tool Governance" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "simonw", "creator_name": "Simon Willison (@simonw)", "creator_url": "https://x.com/simonw", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 93, "installs_count": 320, "agent_pulls_count": 715, "gravity_score": 2585.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_54ec0586a614", "slug": "skill-playwright-visual-qa", "asset_type": "skill", "name": "Playwright Multi-Viewport DOM & Visual Overflow QA Playbook", "tagline": "Run automated headless Playwright verification across desktop (`1440x900`) and mobile (`375x812`) viewports: assert HTTP 200, zero console errors, and zero", "description_md": "### Overview & Capability Scope\n**Playwright Multi-Viewport DOM & Visual Overflow QA Playbook** (`skill-playwright-visual-qa`) is a verified, portable `SKILL.md` playbook curated with **@playwrightweb** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nRun automated headless Playwright verification across desktop (`1440x900`) and mobile (`375x812`) viewports: assert HTTP 200, zero console errors, and zero horizontal overflow.\n\n### Required Stack & Tooling\n- **Primary Stack:** Playwright, Visual QA, Responsive Layout, E2E Testing\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/microsoft/playwright-mcp](https://github.com/microsoft/playwright-mcp)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-playwright-visual-qa\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-playwright-visual-qa/SKILL.md` or `~/.claude/skills/skill-playwright-visual-qa/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx playwright test\nnpx playwright show-report\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "playwright", "visual-qa", "responsive-layout", "e2e-testing", "skill-md", "engineering" ], "source_url": "https://github.com/microsoft/playwright-mcp", "repo_url": "https://github.com/microsoft/playwright-mcp", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-playwright-visual-qa", "skill_markdown": "---\nname: playwright-visual-qa\ndescription: Run automated headless Playwright verification across desktop (`1440x900`) and mobile (`375x812`) viewports: assert HTTP 200, zero console errors, and zero horizontal overflow. Use when working with Playwright, Visual QA, Responsive Layout, E2E Testing.\ntriggers:\n - playwright visual qa\n - playwright\n - visual qa\n---\n\n# Playwright Multi-Viewport DOM & Visual Overflow QA Playbook\n\n## 1. When to Use This Skill\nLoad `skill-playwright-visual-qa` whenever an engineering or operations task involves **Playwright, Visual QA, Responsive Layout, E2E Testing**. Specifically:\n- Run automated headless Playwright verification across desktop (`1440x900`) and mobile (`375x812`) viewports: assert HTTP 200, zero console errors, and zero horizontal overflow.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Playwright, Visual QA, Responsive Layout, E2E Testing\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/microsoft/playwright-mcp`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Playwright**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx playwright test\nnpx playwright show-report\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Playwright Multi-Viewport DOM & Visual Overflow QA Playbook (skill-playwright-visual-qa) playbook. First inspect our current configuration and codebase for Playwright, Visual QA, Responsive Layout, E2E Testing. Then propose a concrete, step-by-step implementation plan following the skill-playwright-visual-qa guardrails, run the non-destructive verification commands (npx playwright test), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Playwright", "Visual QA", "Responsive Layout" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "playwrightweb", "creator_name": "Microsoft Playwright", "creator_url": "https://x.com/playwrightweb", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 92, "installs_count": 316, "agent_pulls_count": 708, "gravity_score": 2556.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_d61b601f8d77", "slug": "skill-visual-sentinel-qa", "asset_type": "skill", "name": "Visual Sentinel Pixel-Diff & Layout Regression Gate Playbook", "tagline": "Automated visual regression gating for CI/CD and autonomous coding agents: capture deterministic baseline screenshots, compute perceptual pixel diffs, and ", "description_md": "### Overview & Capability Scope\n**Visual Sentinel Pixel-Diff & Layout Regression Gate Playbook** (`skill-visual-sentinel-qa`) is a verified, portable `SKILL.md` playbook curated with **@iraiansar** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nAutomated visual regression gating for CI/CD and autonomous coding agents: capture deterministic baseline screenshots, compute perceptual pixel diffs, and block broken UI PRs.\n\n### Required Stack & Tooling\n- **Primary Stack:** Visual Sentinel, Playwright, CI/CD Visual Diff, UI QA\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/iraiansar/visual-sentinel](https://github.com/iraiansar/visual-sentinel)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-visual-sentinel-qa\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-visual-sentinel-qa/SKILL.md` or `~/.claude/skills/skill-visual-sentinel-qa/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx playwright test --update-snapshots\nnpx visual-sentinel compare\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "visual-sentinel", "playwright", "ci/cd-visual-diff", "ui-qa", "skill-md", "engineering" ], "source_url": "https://github.com/iraiansar/visual-sentinel", "repo_url": "https://github.com/iraiansar/visual-sentinel", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-visual-sentinel-qa", "skill_markdown": "---\nname: visual-sentinel-qa\ndescription: Automated visual regression gating for CI/CD and autonomous coding agents: capture deterministic baseline screenshots, compute perceptual pixel diffs, and block broken UI PRs. Use when working with Visual Sentinel, Playwright, CI/CD Visual Diff, UI QA.\ntriggers:\n - visual sentinel qa\n - visual sentinel\n - playwright\n---\n\n# Visual Sentinel Pixel-Diff & Layout Regression Gate Playbook\n\n## 1. When to Use This Skill\nLoad `skill-visual-sentinel-qa` whenever an engineering or operations task involves **Visual Sentinel, Playwright, CI/CD Visual Diff, UI QA**. Specifically:\n- Automated visual regression gating for CI/CD and autonomous coding agents: capture deterministic baseline screenshots, compute perceptual pixel diffs, and block broken UI PRs.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Visual Sentinel, Playwright, CI/CD Visual Diff, UI QA\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/iraiansar/visual-sentinel`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Visual Sentinel**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx playwright test --update-snapshots\nnpx visual-sentinel compare\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Visual Sentinel Pixel-Diff & Layout Regression Gate Playbook (skill-visual-sentinel-qa) playbook. First inspect our current configuration and codebase for Visual Sentinel, Playwright, CI/CD Visual Diff, UI QA. Then propose a concrete, step-by-step implementation plan following the skill-visual-sentinel-qa guardrails, run the non-destructive verification commands (npx playwright test --update-snapshots), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Visual Sentinel", "Playwright", "CI/CD Visual Diff" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "iraiansar", "creator_name": "Irai Ansar (@iraiansar)", "creator_url": "https://x.com/iraiansar", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 91, "installs_count": 312, "agent_pulls_count": 701, "gravity_score": 2527.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_d62f14d1ba60", "slug": "skill-web-perf", "asset_type": "skill", "name": "Core Web Vitals (LCP, INP, CLS) & Lighthouse 100 Audit Skill", "tagline": "Audit and optimize website loading and interaction performance: eliminate render-blocking CSS/JS, preload LCP hero assets, prevent layout shifts, and reduc", "description_md": "### Overview & Capability Scope\n**Core Web Vitals (LCP, INP, CLS) & Lighthouse 100 Audit Skill** (`skill-web-perf`) is a verified, portable `SKILL.md` playbook curated with **@cloudflare** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nAudit and optimize website loading and interaction performance: eliminate render-blocking CSS/JS, preload LCP hero assets, prevent layout shifts, and reduce main-thread TBT.\n\n### Required Stack & Tooling\n- **Primary Stack:** Core Web Vitals, Lighthouse, LCP / INP / CLS, Web Performance\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/GoogleChrome/lighthouse](https://github.com/GoogleChrome/lighthouse)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-web-perf\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-web-perf/SKILL.md` or `~/.claude/skills/skill-web-perf/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx lighthouse https://example.com --output=json --chrome-flags='--headless'\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "core-web-vitals", "lighthouse", "lcp-/-inp-/-cls", "web-performance", "skill-md", "engineering" ], "source_url": "https://github.com/GoogleChrome/lighthouse", "repo_url": "https://github.com/GoogleChrome/lighthouse", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-web-perf", "skill_markdown": "---\nname: web-perf\ndescription: Audit and optimize website loading and interaction performance: eliminate render-blocking CSS/JS, preload LCP hero assets, prevent layout shifts, and reduce main-thread TBT. Use when working with Core Web Vitals, Lighthouse, LCP / INP / CLS, Web Performance.\ntriggers:\n - web perf\n - core web vitals\n - lighthouse\n---\n\n# Core Web Vitals (LCP, INP, CLS) & Lighthouse 100 Audit Skill\n\n## 1. When to Use This Skill\nLoad `skill-web-perf` whenever an engineering or operations task involves **Core Web Vitals, Lighthouse, LCP / INP / CLS, Web Performance**. Specifically:\n- Audit and optimize website loading and interaction performance: eliminate render-blocking CSS/JS, preload LCP hero assets, prevent layout shifts, and reduce main-thread TBT.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Core Web Vitals, Lighthouse, LCP / INP / CLS, Web Performance\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/GoogleChrome/lighthouse`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Core Web Vitals**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx lighthouse https://example.com --output=json --chrome-flags='--headless'\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Core Web Vitals (LCP, INP, CLS) & Lighthouse 100 Audit Skill (skill-web-perf) playbook. First inspect our current configuration and codebase for Core Web Vitals, Lighthouse, LCP / INP / CLS, Web Performance. Then propose a concrete, step-by-step implementation plan following the skill-web-perf guardrails, run the non-destructive verification commands (npx lighthouse https://example.com --output=json --chrome-flags='--headless'), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Core Web Vitals", "Lighthouse", "LCP / INP / CLS" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "cloudflare", "creator_name": "Cloudflare Developers", "creator_url": "https://x.com/cloudflare", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 90, "installs_count": 308, "agent_pulls_count": 694, "gravity_score": 2498.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_50dfb8c895f1", "slug": "skill-github-actions-ci", "asset_type": "skill", "name": "GitHub Actions Matrix Caching, OIDC & Fast CI Pipeline Skill", "tagline": "Cut GitHub Actions CI runtimes by 60%: configure pnpm/Cargo/Docker layer caching, concurrency cancel-in-progress groups, and keyless OIDC cloud deployments", "description_md": "### Overview & Capability Scope\n**GitHub Actions Matrix Caching, OIDC & Fast CI Pipeline Skill** (`skill-github-actions-ci`) is a verified, portable `SKILL.md` playbook curated with **@steipete** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nCut GitHub Actions CI runtimes by 60%: configure pnpm/Cargo/Docker layer caching, concurrency cancel-in-progress groups, and keyless OIDC cloud deployments.\n\n### Required Stack & Tooling\n- **Primary Stack:** GitHub Actions, CI/CD Optimization, OIDC Auth, Build Caching\n- **Category:** `engineering`\n- **Source & Reference:** [https://docs.github.com/en/actions](https://docs.github.com/en/actions)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-github-actions-ci\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-github-actions-ci/SKILL.md` or `~/.claude/skills/skill-github-actions-ci/SKILL.md`.\n\n### Verification Commands\n```bash\ngh run list --limit 5\ngh workflow view\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "github-actions", "ci/cd-optimization", "oidc-auth", "build-caching", "skill-md", "engineering" ], "source_url": "https://docs.github.com/en/actions", "repo_url": "https://docs.github.com/en/actions", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-github-actions-ci", "skill_markdown": "---\nname: github-actions-ci\ndescription: Cut GitHub Actions CI runtimes by 60%: configure pnpm/Cargo/Docker layer caching, concurrency cancel-in-progress groups, and keyless OIDC cloud deployments. Use when working with GitHub Actions, CI/CD Optimization, OIDC Auth, Build Caching.\ntriggers:\n - github actions ci\n - github actions\n - ci/cd optimization\n---\n\n# GitHub Actions Matrix Caching, OIDC & Fast CI Pipeline Skill\n\n## 1. When to Use This Skill\nLoad `skill-github-actions-ci` whenever an engineering or operations task involves **GitHub Actions, CI/CD Optimization, OIDC Auth, Build Caching**. Specifically:\n- Cut GitHub Actions CI runtimes by 60%: configure pnpm/Cargo/Docker layer caching, concurrency cancel-in-progress groups, and keyless OIDC cloud deployments.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** GitHub Actions, CI/CD Optimization, OIDC Auth, Build Caching\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://docs.github.com/en/actions`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **GitHub Actions**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\ngh run list --limit 5\ngh workflow view\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the GitHub Actions Matrix Caching, OIDC & Fast CI Pipeline Skill (skill-github-actions-ci) playbook. First inspect our current configuration and codebase for GitHub Actions, CI/CD Optimization, OIDC Auth, Build Caching. Then propose a concrete, step-by-step implementation plan following the skill-github-actions-ci guardrails, run the non-destructive verification commands (gh run list --limit 5), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "GitHub Actions", "CI/CD Optimization", "OIDC Auth" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "steipete", "creator_name": "Peter Steinberger (@steipete)", "creator_url": "https://x.com/steipete", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 89, "installs_count": 304, "agent_pulls_count": 687, "gravity_score": 2469.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_a3c2e1b28794", "slug": "skill-docker-multistage", "asset_type": "skill", "name": "Docker Multi-Stage Distroless Builds & Image Hardening Skill", "tagline": "Shrink container images by 80% and eliminate OS-level CVEs using multi-stage BuildKit Dockerfiles, non-root distroless runtime images, and Trivy vulnerabil", "description_md": "### Overview & Capability Scope\n**Docker Multi-Stage Distroless Builds & Image Hardening Skill** (`skill-docker-multistage`) is a verified, portable `SKILL.md` playbook curated with **@steipete** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nShrink container images by 80% and eliminate OS-level CVEs using multi-stage BuildKit Dockerfiles, non-root distroless runtime images, and Trivy vulnerability scans.\n\n### Required Stack & Tooling\n- **Primary Stack:** Docker BuildKit, Distroless, Container Security, Trivy\n- **Category:** `security-ops`\n- **Source & Reference:** [https://docs.docker.com/build/building/multi-stage/](https://docs.docker.com/build/building/multi-stage/)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-docker-multistage\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-docker-multistage/SKILL.md` or `~/.claude/skills/skill-docker-multistage/SKILL.md`.\n\n### Verification Commands\n```bash\ndocker buildx build --target runtime -t app:latest .\ntrivy image app:latest\n```\n", "category": "security-ops", "subcategories": [ "security-ops", "skills" ], "tags": [ "docker-buildkit", "distroless", "container-security", "trivy", "skill-md", "security-ops" ], "source_url": "https://docs.docker.com/build/building/multi-stage/", "repo_url": "https://docs.docker.com/build/building/multi-stage/", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-docker-multistage", "skill_markdown": "---\nname: docker-multistage\ndescription: Shrink container images by 80% and eliminate OS-level CVEs using multi-stage BuildKit Dockerfiles, non-root distroless runtime images, and Trivy vulnerability scans. Use when working with Docker BuildKit, Distroless, Container Security, Trivy.\ntriggers:\n - docker multistage\n - docker buildkit\n - distroless\n---\n\n# Docker Multi-Stage Distroless Builds & Image Hardening Skill\n\n## 1. When to Use This Skill\nLoad `skill-docker-multistage` whenever an engineering or operations task involves **Docker BuildKit, Distroless, Container Security, Trivy**. Specifically:\n- Shrink container images by 80% and eliminate OS-level CVEs using multi-stage BuildKit Dockerfiles, non-root distroless runtime images, and Trivy vulnerability scans.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Docker BuildKit, Distroless, Container Security, Trivy\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://docs.docker.com/build/building/multi-stage/`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Docker BuildKit**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\ndocker buildx build --target runtime -t app:latest .\ntrivy image app:latest\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Docker Multi-Stage Distroless Builds & Image Hardening Skill (skill-docker-multistage) playbook. First inspect our current configuration and codebase for Docker BuildKit, Distroless, Container Security, Trivy. Then propose a concrete, step-by-step implementation plan following the skill-docker-multistage guardrails, run the non-destructive verification commands (docker buildx build --target runtime -t app:latest .), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Docker BuildKit", "Distroless", "Container Security" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "steipete", "creator_name": "Peter Steinberger (@steipete)", "creator_url": "https://x.com/steipete", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 88, "installs_count": 300, "agent_pulls_count": 680, "gravity_score": 2440.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_9eb980c64b54", "slug": "skill-sentry-root-cause", "asset_type": "skill", "name": "Sentry Stack-Trace Triage, Source Maps & Root-Cause Skill", "tagline": "Query production Sentry issues, resolve minified JavaScript/TypeScript stack traces via source maps, correlate release commits, and generate minimal reprod", "description_md": "### Overview & Capability Scope\n**Sentry Stack-Trace Triage, Source Maps & Root-Cause Skill** (`skill-sentry-root-cause`) is a verified, portable `SKILL.md` playbook curated with **@getsentry** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nQuery production Sentry issues, resolve minified JavaScript/TypeScript stack traces via source maps, correlate release commits, and generate minimal reproduction tests.\n\n### Required Stack & Tooling\n- **Primary Stack:** Sentry MCP, Source Maps, Incident Triage, Error Monitoring\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/getsentry/sentry-mcp](https://github.com/getsentry/sentry-mcp)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-sentry-root-cause\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-sentry-root-cause/SKILL.md` or `~/.claude/skills/skill-sentry-root-cause/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx @sentry/wizard@latest -i sourcemaps\nsentry-cli releases list\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "sentry-mcp", "source-maps", "incident-triage", "error-monitoring", "skill-md", "engineering" ], "source_url": "https://github.com/getsentry/sentry-mcp", "repo_url": "https://github.com/getsentry/sentry-mcp", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-sentry-root-cause", "skill_markdown": "---\nname: sentry-root-cause\ndescription: Query production Sentry issues, resolve minified JavaScript/TypeScript stack traces via source maps, correlate release commits, and generate minimal reproduction tests. Use when working with Sentry MCP, Source Maps, Incident Triage, Error Monitoring.\ntriggers:\n - sentry root cause\n - sentry mcp\n - source maps\n---\n\n# Sentry Stack-Trace Triage, Source Maps & Root-Cause Skill\n\n## 1. When to Use This Skill\nLoad `skill-sentry-root-cause` whenever an engineering or operations task involves **Sentry MCP, Source Maps, Incident Triage, Error Monitoring**. Specifically:\n- Query production Sentry issues, resolve minified JavaScript/TypeScript stack traces via source maps, correlate release commits, and generate minimal reproduction tests.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Sentry MCP, Source Maps, Incident Triage, Error Monitoring\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/getsentry/sentry-mcp`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Sentry MCP**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx @sentry/wizard@latest -i sourcemaps\nsentry-cli releases list\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Sentry Stack-Trace Triage, Source Maps & Root-Cause Skill (skill-sentry-root-cause) playbook. First inspect our current configuration and codebase for Sentry MCP, Source Maps, Incident Triage, Error Monitoring. Then propose a concrete, step-by-step implementation plan following the skill-sentry-root-cause guardrails, run the non-destructive verification commands (npx @sentry/wizard@latest -i sourcemaps), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Sentry MCP", "Source Maps", "Incident Triage" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "getsentry", "creator_name": "Sentry Engineering", "creator_url": "https://x.com/getsentry", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 87, "installs_count": 296, "agent_pulls_count": 673, "gravity_score": 2411.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_c9ed8e72c4e5", "slug": "skill-opentelemetry-tracing", "asset_type": "skill", "name": "OpenTelemetry Distributed Tracing & Span Context Skill", "tagline": "Instrument Node.js, Edge Workers, and Python services with OpenTelemetry W3C TraceContext propagation, structured JSON logs, and OTLP collector exporters.", "description_md": "### Overview & Capability Scope\n**OpenTelemetry Distributed Tracing & Span Context Skill** (`skill-opentelemetry-tracing`) is a verified, portable `SKILL.md` playbook curated with **@getsentry** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nInstrument Node.js, Edge Workers, and Python services with OpenTelemetry W3C TraceContext propagation, structured JSON logs, and OTLP collector exporters.\n\n### Required Stack & Tooling\n- **Primary Stack:** OpenTelemetry, W3C TraceContext, OTLP, Observability\n- **Category:** `engineering`\n- **Source & Reference:** [https://opentelemetry.io/docs/](https://opentelemetry.io/docs/)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-opentelemetry-tracing\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-opentelemetry-tracing/SKILL.md` or `~/.claude/skills/skill-opentelemetry-tracing/SKILL.md`.\n\n### Verification Commands\n```bash\nnpm i @opentelemetry/api @opentelemetry/sdk-node\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "opentelemetry", "w3c-tracecontext", "otlp", "observability", "skill-md", "engineering" ], "source_url": "https://opentelemetry.io/docs/", "repo_url": "https://opentelemetry.io/docs/", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-opentelemetry-tracing", "skill_markdown": "---\nname: opentelemetry-tracing\ndescription: Instrument Node.js, Edge Workers, and Python services with OpenTelemetry W3C TraceContext propagation, structured JSON logs, and OTLP collector exporters. Use when working with OpenTelemetry, W3C TraceContext, OTLP, Observability.\ntriggers:\n - opentelemetry tracing\n - opentelemetry\n - w3c tracecontext\n---\n\n# OpenTelemetry Distributed Tracing & Span Context Skill\n\n## 1. When to Use This Skill\nLoad `skill-opentelemetry-tracing` whenever an engineering or operations task involves **OpenTelemetry, W3C TraceContext, OTLP, Observability**. Specifically:\n- Instrument Node.js, Edge Workers, and Python services with OpenTelemetry W3C TraceContext propagation, structured JSON logs, and OTLP collector exporters.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** OpenTelemetry, W3C TraceContext, OTLP, Observability\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://opentelemetry.io/docs/`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **OpenTelemetry**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpm i @opentelemetry/api @opentelemetry/sdk-node\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the OpenTelemetry Distributed Tracing & Span Context Skill (skill-opentelemetry-tracing) playbook. First inspect our current configuration and codebase for OpenTelemetry, W3C TraceContext, OTLP, Observability. Then propose a concrete, step-by-step implementation plan following the skill-opentelemetry-tracing guardrails, run the non-destructive verification commands (npm i @opentelemetry/api @opentelemetry/sdk-node), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "OpenTelemetry", "W3C TraceContext", "OTLP" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "getsentry", "creator_name": "Sentry Engineering", "creator_url": "https://x.com/getsentry", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 86, "installs_count": 292, "agent_pulls_count": 666, "gravity_score": 2382.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_ef21f9e99f26", "slug": "skill-owasp-api-security", "asset_type": "skill", "name": "OWASP Top-10 API Security, AuthZ & Secret Hygiene Audit Skill", "tagline": "Systematic security audit for REST and GraphQL APIs: detect Broken Object Level Authorization (BOLA/IDOR), mass assignment, SSRF, JWT algorithm confusion, ", "description_md": "### Overview & Capability Scope\n**OWASP Top-10 API Security, AuthZ & Secret Hygiene Audit Skill** (`skill-owasp-api-security`) is a verified, portable `SKILL.md` playbook curated with **@simonw** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nSystematic security audit for REST and GraphQL APIs: detect Broken Object Level Authorization (BOLA/IDOR), mass assignment, SSRF, JWT algorithm confusion, and leaked secrets.\n\n### Required Stack & Tooling\n- **Primary Stack:** OWASP API Top 10, BOLA / IDOR, JWT Security, SSRF Prevention\n- **Category:** `security-ops`\n- **Source & Reference:** [https://owasp.org/www-project-api-security/](https://owasp.org/www-project-api-security/)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-owasp-api-security\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-owasp-api-security/SKILL.md` or `~/.claude/skills/skill-owasp-api-security/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx gitleaks detect --source .\nnpm audit --omit=dev\n```\n", "category": "security-ops", "subcategories": [ "security-ops", "skills" ], "tags": [ "owasp-api-top-10", "bola-/-idor", "jwt-security", "ssrf-prevention", "skill-md", "security-ops" ], "source_url": "https://owasp.org/www-project-api-security/", "repo_url": "https://owasp.org/www-project-api-security/", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-owasp-api-security", "skill_markdown": "---\nname: owasp-api-security\ndescription: Systematic security audit for REST and GraphQL APIs: detect Broken Object Level Authorization (BOLA/IDOR), mass assignment, SSRF, JWT algorithm confusion, and leaked secrets. Use when working with OWASP API Top 10, BOLA / IDOR, JWT Security, SSRF Prevention.\ntriggers:\n - owasp api security\n - owasp api top 10\n - bola / idor\n---\n\n# OWASP Top-10 API Security, AuthZ & Secret Hygiene Audit Skill\n\n## 1. When to Use This Skill\nLoad `skill-owasp-api-security` whenever an engineering or operations task involves **OWASP API Top 10, BOLA / IDOR, JWT Security, SSRF Prevention**. Specifically:\n- Systematic security audit for REST and GraphQL APIs: detect Broken Object Level Authorization (BOLA/IDOR), mass assignment, SSRF, JWT algorithm confusion, and leaked secrets.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** OWASP API Top 10, BOLA / IDOR, JWT Security, SSRF Prevention\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://owasp.org/www-project-api-security/`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **OWASP API Top 10**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx gitleaks detect --source .\nnpm audit --omit=dev\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the OWASP Top-10 API Security, AuthZ & Secret Hygiene Audit Skill (skill-owasp-api-security) playbook. First inspect our current configuration and codebase for OWASP API Top 10, BOLA / IDOR, JWT Security, SSRF Prevention. Then propose a concrete, step-by-step implementation plan following the skill-owasp-api-security guardrails, run the non-destructive verification commands (npx gitleaks detect --source .), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "OWASP API Top 10", "BOLA / IDOR", "JWT Security" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "simonw", "creator_name": "Simon Willison (@simonw)", "creator_url": "https://x.com/simonw", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 85, "installs_count": 288, "agent_pulls_count": 659, "gravity_score": 2353.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_38bd4a128329", "slug": "skill-dependency-cve-patching", "asset_type": "skill", "name": "Automated Dependency CVE Triage & Semver Lockfile Upgrade Skill", "tagline": "Audit `package-lock.json`, `pnpm-lock.yaml`, and `Cargo.lock` for reachable CVEs, test minimal semver patches against the full test suite, and open verifie", "description_md": "### Overview & Capability Scope\n**Automated Dependency CVE Triage & Semver Lockfile Upgrade Skill** (`skill-dependency-cve-patching`) is a verified, portable `SKILL.md` playbook curated with **@kentcdodds** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nAudit `package-lock.json`, `pnpm-lock.yaml`, and `Cargo.lock` for reachable CVEs, test minimal semver patches against the full test suite, and open verified PRs.\n\n### Required Stack & Tooling\n- **Primary Stack:** CVE Triage, npm / pnpm, Supply Chain Security, Automated PRs\n- **Category:** `security-ops`\n- **Source & Reference:** [https://docs.github.com/en/code-security/dependabot](https://docs.github.com/en/code-security/dependabot)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-dependency-cve-patching\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-dependency-cve-patching/SKILL.md` or `~/.claude/skills/skill-dependency-cve-patching/SKILL.md`.\n\n### Verification Commands\n```bash\nnpm audit --json\nnpx npm-check-updates --target minor\n```\n", "category": "security-ops", "subcategories": [ "security-ops", "skills" ], "tags": [ "cve-triage", "npm-/-pnpm", "supply-chain-security", "automated-prs", "skill-md", "security-ops" ], "source_url": "https://docs.github.com/en/code-security/dependabot", "repo_url": "https://docs.github.com/en/code-security/dependabot", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-dependency-cve-patching", "skill_markdown": "---\nname: dependency-cve-patching\ndescription: Audit `package-lock.json`, `pnpm-lock.yaml`, and `Cargo.lock` for reachable CVEs, test minimal semver patches against the full test suite, and open verified PRs. Use when working with CVE Triage, npm / pnpm, Supply Chain Security, Automated PRs.\ntriggers:\n - dependency cve patching\n - cve triage\n - npm / pnpm\n---\n\n# Automated Dependency CVE Triage & Semver Lockfile Upgrade Skill\n\n## 1. When to Use This Skill\nLoad `skill-dependency-cve-patching` whenever an engineering or operations task involves **CVE Triage, npm / pnpm, Supply Chain Security, Automated PRs**. Specifically:\n- Audit `package-lock.json`, `pnpm-lock.yaml`, and `Cargo.lock` for reachable CVEs, test minimal semver patches against the full test suite, and open verified PRs.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** CVE Triage, npm / pnpm, Supply Chain Security, Automated PRs\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://docs.github.com/en/code-security/dependabot`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **CVE Triage**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpm audit --json\nnpx npm-check-updates --target minor\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Automated Dependency CVE Triage & Semver Lockfile Upgrade Skill (skill-dependency-cve-patching) playbook. First inspect our current configuration and codebase for CVE Triage, npm / pnpm, Supply Chain Security, Automated PRs. Then propose a concrete, step-by-step implementation plan following the skill-dependency-cve-patching guardrails, run the non-destructive verification commands (npm audit --json), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "CVE Triage", "npm / pnpm", "Supply Chain Security" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "kentcdodds", "creator_name": "Kent C. Dodds (@kentcdodds)", "creator_url": "https://x.com/kentcdodds", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 84, "installs_count": 284, "agent_pulls_count": 652, "gravity_score": 2324.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_2ebb222fa1f5", "slug": "skill-slack-api", "asset_type": "skill", "name": "Slack Web API OAuth Scopes, Rate Limits & Event Subscriptions", "tagline": "Choose the right Slack Web API methods (`chat.postMessage`, `conversations.history`, `users.info`), configure least-privilege bot token scopes, and verify ", "description_md": "### Overview & Capability Scope\n**Slack Web API OAuth Scopes, Rate Limits & Event Subscriptions** (`skill-slack-api`) is a verified, portable `SKILL.md` playbook curated with **@linear** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nChoose the right Slack Web API methods (`chat.postMessage`, `conversations.history`, `users.info`), configure least-privilege bot token scopes, and verify webhook signatures.\n\n### Required Stack & Tooling\n- **Primary Stack:** Slack Web API, Bolt SDK, OAuth Scopes, Events API\n- **Category:** `operations`\n- **Source & Reference:** [https://docs.slack.dev/apis/web-api](https://docs.slack.dev/apis/web-api)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-slack-api\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-slack-api/SKILL.md` or `~/.claude/skills/skill-slack-api/SKILL.md`.\n\n### Verification Commands\n```bash\nslack run\nslack trigger list\n```\n", "category": "operations", "subcategories": [ "operations", "skills" ], "tags": [ "slack-web-api", "bolt-sdk", "oauth-scopes", "events-api", "skill-md", "operations" ], "source_url": "https://docs.slack.dev/apis/web-api", "repo_url": "https://docs.slack.dev/apis/web-api", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-slack-api", "skill_markdown": "---\nname: slack-api\ndescription: Choose the right Slack Web API methods (`chat.postMessage`, `conversations.history`, `users.info`), configure least-privilege bot token scopes, and verify webhook signatures. Use when working with Slack Web API, Bolt SDK, OAuth Scopes, Events API.\ntriggers:\n - slack api\n - slack web api\n - bolt sdk\n---\n\n# Slack Web API OAuth Scopes, Rate Limits & Event Subscriptions\n\n## 1. When to Use This Skill\nLoad `skill-slack-api` whenever an engineering or operations task involves **Slack Web API, Bolt SDK, OAuth Scopes, Events API**. Specifically:\n- Choose the right Slack Web API methods (`chat.postMessage`, `conversations.history`, `users.info`), configure least-privilege bot token scopes, and verify webhook signatures.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Slack Web API, Bolt SDK, OAuth Scopes, Events API\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://docs.slack.dev/apis/web-api`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Slack Web API**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nslack run\nslack trigger list\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Slack Web API OAuth Scopes, Rate Limits & Event Subscriptions (skill-slack-api) playbook. First inspect our current configuration and codebase for Slack Web API, Bolt SDK, OAuth Scopes, Events API. Then propose a concrete, step-by-step implementation plan following the skill-slack-api guardrails, run the non-destructive verification commands (slack run), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Slack Web API", "Bolt SDK", "OAuth Scopes" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "linear", "creator_name": "Linear Engineering", "creator_url": "https://x.com/linear", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 94, "installs_count": 276, "agent_pulls_count": 638, "gravity_score": 2296.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_e2e9c7dde69a", "slug": "skill-block-kit", "asset_type": "skill", "name": "Slack Block Kit Interactive Modals, Home Tabs & Action Payloads", "tagline": "Design and validate Slack Block Kit JSON payloads for interactive approval cards, slash-command modals, Home tabs, and stateful button callbacks within the", "description_md": "### Overview & Capability Scope\n**Slack Block Kit Interactive Modals, Home Tabs & Action Payloads** (`skill-block-kit`) is a verified, portable `SKILL.md` playbook curated with **@linear** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nDesign and validate Slack Block Kit JSON payloads for interactive approval cards, slash-command modals, Home tabs, and stateful button callbacks within the 50-block limit.\n\n### Required Stack & Tooling\n- **Primary Stack:** Slack Block Kit, Interactive Modals, Slack Web API, Approval Cards\n- **Category:** `operations`\n- **Source & Reference:** [https://docs.slack.dev/block-kit](https://docs.slack.dev/block-kit)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-block-kit\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-block-kit/SKILL.md` or `~/.claude/skills/skill-block-kit/SKILL.md`.\n\n### Verification Commands\n```bash\nslack manifest validate\ncurl -X POST https://slack.com/api/chat.postMessage\n```\n", "category": "operations", "subcategories": [ "operations", "skills" ], "tags": [ "slack-block-kit", "interactive-modals", "slack-web-api", "approval-cards", "skill-md", "operations" ], "source_url": "https://docs.slack.dev/block-kit", "repo_url": "https://docs.slack.dev/block-kit", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-block-kit", "skill_markdown": "---\nname: block-kit\ndescription: Design and validate Slack Block Kit JSON payloads for interactive approval cards, slash-command modals, Home tabs, and stateful button callbacks within the 50-block limit. Use when working with Slack Block Kit, Interactive Modals, Slack Web API, Approval Cards.\ntriggers:\n - block kit\n - slack block kit\n - interactive modals\n---\n\n# Slack Block Kit Interactive Modals, Home Tabs & Action Payloads\n\n## 1. When to Use This Skill\nLoad `skill-block-kit` whenever an engineering or operations task involves **Slack Block Kit, Interactive Modals, Slack Web API, Approval Cards**. Specifically:\n- Design and validate Slack Block Kit JSON payloads for interactive approval cards, slash-command modals, Home tabs, and stateful button callbacks within the 50-block limit.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Slack Block Kit, Interactive Modals, Slack Web API, Approval Cards\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://docs.slack.dev/block-kit`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Slack Block Kit**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nslack manifest validate\ncurl -X POST https://slack.com/api/chat.postMessage\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Slack Block Kit Interactive Modals, Home Tabs & Action Payloads (skill-block-kit) playbook. First inspect our current configuration and codebase for Slack Block Kit, Interactive Modals, Slack Web API, Approval Cards. Then propose a concrete, step-by-step implementation plan following the skill-block-kit guardrails, run the non-destructive verification commands (slack manifest validate), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Slack Block Kit", "Interactive Modals", "Slack Web API" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "linear", "creator_name": "Linear Engineering", "creator_url": "https://x.com/linear", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 83, "installs_count": 280, "agent_pulls_count": 645, "gravity_score": 2295.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_9e5cf7c42e5f", "slug": "skill-database-query", "asset_type": "skill", "name": "Notion Database Schema Inspection, Filtered Queries & Rollups", "tagline": "Query Notion databases deterministically by ID or title: resolve property schemas (`status`, `select`, `relation`, `date`), apply compound filters, and for", "description_md": "### Overview & Capability Scope\n**Notion Database Schema Inspection, Filtered Queries & Rollups** (`skill-database-query`) is a verified, portable `SKILL.md` playbook curated with **@mckaywrigley** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nQuery Notion databases deterministically by ID or title: resolve property schemas (`status`, `select`, `relation`, `date`), apply compound filters, and format structured tables.\n\n### Required Stack & Tooling\n- **Primary Stack:** Notion MCP, Database Queries, Structured Knowledge, Workspace Ops\n- **Category:** `chief-of-staff`\n- **Source & Reference:** [https://github.com/makenotion/notion-mcp-server](https://github.com/makenotion/notion-mcp-server)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-database-query\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-database-query/SKILL.md` or `~/.claude/skills/skill-database-query/SKILL.md`.\n\n### Verification Commands\n```bash\nclaude mcp add --transport http notion https://mcp.notion.com/mcp\n```\n", "category": "chief-of-staff", "subcategories": [ "chief-of-staff", "skills" ], "tags": [ "notion-mcp", "database-queries", "structured-knowledge", "workspace-ops", "skill-md", "chief-of-staff" ], "source_url": "https://github.com/makenotion/notion-mcp-server", "repo_url": "https://github.com/makenotion/notion-mcp-server", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-database-query", "skill_markdown": "---\nname: database-query\ndescription: Query Notion databases deterministically by ID or title: resolve property schemas (`status`, `select`, `relation`, `date`), apply compound filters, and format structured tables. Use when working with Notion MCP, Database Queries, Structured Knowledge, Workspace Ops.\ntriggers:\n - database query\n - notion mcp\n - database queries\n---\n\n# Notion Database Schema Inspection, Filtered Queries & Rollups\n\n## 1. When to Use This Skill\nLoad `skill-database-query` whenever an engineering or operations task involves **Notion MCP, Database Queries, Structured Knowledge, Workspace Ops**. Specifically:\n- Query Notion databases deterministically by ID or title: resolve property schemas (`status`, `select`, `relation`, `date`), apply compound filters, and format structured tables.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Notion MCP, Database Queries, Structured Knowledge, Workspace Ops\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/makenotion/notion-mcp-server`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Notion MCP**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nclaude mcp add --transport http notion https://mcp.notion.com/mcp\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Notion Database Schema Inspection, Filtered Queries & Rollups (skill-database-query) playbook. First inspect our current configuration and codebase for Notion MCP, Database Queries, Structured Knowledge, Workspace Ops. Then propose a concrete, step-by-step implementation plan following the skill-database-query guardrails, run the non-destructive verification commands (claude mcp add --transport http notion https://mcp.notion.com/mcp), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Notion MCP", "Database Queries", "Structured Knowledge" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "mckaywrigley", "creator_name": "Mckay Wrigley (@mckaywrigley)", "creator_url": "https://x.com/mckaywrigley", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 93, "installs_count": 272, "agent_pulls_count": 631, "gravity_score": 2267.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_09d53d481bca", "slug": "skill-spec-to-implementation", "asset_type": "skill", "name": "Product Spec to Phased Engineering Implementation Plan Skill", "tagline": "Transform high-level product requirements and RFCs into dependency-ordered engineering tasks with explicit acceptance criteria, database migrations, and QA", "description_md": "### Overview & Capability Scope\n**Product Spec to Phased Engineering Implementation Plan Skill** (`skill-spec-to-implementation`) is a verified, portable `SKILL.md` playbook curated with **@lennysan** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nTransform high-level product requirements and RFCs into dependency-ordered engineering tasks with explicit acceptance criteria, database migrations, and QA verification steps.\n\n### Required Stack & Tooling\n- **Primary Stack:** Product Specs, Task Decomposition, Linear / Notion, Acceptance Criteria\n- **Category:** `chief-of-staff`\n- **Source & Reference:** [https://github.com/makenotion/notion-mcp-server](https://github.com/makenotion/notion-mcp-server)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-spec-to-implementation\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-spec-to-implementation/SKILL.md` or `~/.claude/skills/skill-spec-to-implementation/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx lightstate inspect skill-spec-to-implementation\n```\n", "category": "chief-of-staff", "subcategories": [ "chief-of-staff", "skills" ], "tags": [ "product-specs", "task-decomposition", "linear-/-notion", "acceptance-criteria", "skill-md", "chief-of-staff" ], "source_url": "https://github.com/makenotion/notion-mcp-server", "repo_url": "https://github.com/makenotion/notion-mcp-server", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-spec-to-implementation", "skill_markdown": "---\nname: spec-to-implementation\ndescription: Transform high-level product requirements and RFCs into dependency-ordered engineering tasks with explicit acceptance criteria, database migrations, and QA verification steps. Use when working with Product Specs, Task Decomposition, Linear / Notion, Acceptance Criteria.\ntriggers:\n - spec to implementation\n - product specs\n - task decomposition\n---\n\n# Product Spec to Phased Engineering Implementation Plan Skill\n\n## 1. When to Use This Skill\nLoad `skill-spec-to-implementation` whenever an engineering or operations task involves **Product Specs, Task Decomposition, Linear / Notion, Acceptance Criteria**. Specifically:\n- Transform high-level product requirements and RFCs into dependency-ordered engineering tasks with explicit acceptance criteria, database migrations, and QA verification steps.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Product Specs, Task Decomposition, Linear / Notion, Acceptance Criteria\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/makenotion/notion-mcp-server`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Product Specs**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx lightstate inspect skill-spec-to-implementation\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Product Spec to Phased Engineering Implementation Plan Skill (skill-spec-to-implementation) playbook. First inspect our current configuration and codebase for Product Specs, Task Decomposition, Linear / Notion, Acceptance Criteria. Then propose a concrete, step-by-step implementation plan following the skill-spec-to-implementation guardrails, run the non-destructive verification commands (npx lightstate inspect skill-spec-to-implementation), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Product Specs", "Task Decomposition", "Linear / Notion" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "lennysan", "creator_name": "Lenny Rachitsky (@lennysan)", "creator_url": "https://x.com/lennysan", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 92, "installs_count": 268, "agent_pulls_count": 624, "gravity_score": 2238.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_233701d3ce8b", "slug": "skill-linear-sprint-hygiene", "asset_type": "skill", "name": "Linear Sprint Triage, Duplicate Detection & Cycle Hygiene Skill", "tagline": "Automate Linear issue triage: detect duplicate bug reports, infer team/project labels from stack traces, flag blocked cycle items, and draft weekly enginee", "description_md": "### Overview & Capability Scope\n**Linear Sprint Triage, Duplicate Detection & Cycle Hygiene Skill** (`skill-linear-sprint-hygiene`) is a verified, portable `SKILL.md` playbook curated with **@linear** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nAutomate Linear issue triage: detect duplicate bug reports, infer team/project labels from stack traces, flag blocked cycle items, and draft weekly engineering release notes.\n\n### Required Stack & Tooling\n- **Primary Stack:** Linear MCP, Issue Triage, Sprint Cycles, Release Notes\n- **Category:** `engineering`\n- **Source & Reference:** [https://linear.app/docs/mcp](https://linear.app/docs/mcp)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-linear-sprint-hygiene\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-linear-sprint-hygiene/SKILL.md` or `~/.claude/skills/skill-linear-sprint-hygiene/SKILL.md`.\n\n### Verification Commands\n```bash\nclaude mcp add --transport http linear https://mcp.linear.app/mcp\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "linear-mcp", "issue-triage", "sprint-cycles", "release-notes", "skill-md", "engineering" ], "source_url": "https://linear.app/docs/mcp", "repo_url": "https://linear.app/docs/mcp", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-linear-sprint-hygiene", "skill_markdown": "---\nname: linear-sprint-hygiene\ndescription: Automate Linear issue triage: detect duplicate bug reports, infer team/project labels from stack traces, flag blocked cycle items, and draft weekly engineering release notes. Use when working with Linear MCP, Issue Triage, Sprint Cycles, Release Notes.\ntriggers:\n - linear sprint hygiene\n - linear mcp\n - issue triage\n---\n\n# Linear Sprint Triage, Duplicate Detection & Cycle Hygiene Skill\n\n## 1. When to Use This Skill\nLoad `skill-linear-sprint-hygiene` whenever an engineering or operations task involves **Linear MCP, Issue Triage, Sprint Cycles, Release Notes**. Specifically:\n- Automate Linear issue triage: detect duplicate bug reports, infer team/project labels from stack traces, flag blocked cycle items, and draft weekly engineering release notes.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Linear MCP, Issue Triage, Sprint Cycles, Release Notes\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://linear.app/docs/mcp`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Linear MCP**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nclaude mcp add --transport http linear https://mcp.linear.app/mcp\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Linear Sprint Triage, Duplicate Detection & Cycle Hygiene Skill (skill-linear-sprint-hygiene) playbook. First inspect our current configuration and codebase for Linear MCP, Issue Triage, Sprint Cycles, Release Notes. Then propose a concrete, step-by-step implementation plan following the skill-linear-sprint-hygiene guardrails, run the non-destructive verification commands (claude mcp add --transport http linear https://mcp.linear.app/mcp), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Linear MCP", "Issue Triage", "Sprint Cycles" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "linear", "creator_name": "Linear Engineering", "creator_url": "https://x.com/linear", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 91, "installs_count": 264, "agent_pulls_count": 617, "gravity_score": 2209.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_79594d35e78f", "slug": "skill-google-sheets", "asset_type": "skill", "name": "Google Sheets Financial Model, KPI Tracker & Formula Playbook", "tagline": "Build structured Google Sheets ledgers and KPI dashboards programmatically: batch value updates, `INDEX/MATCH` & `QUERY` formulas, conditional formatting, ", "description_md": "### Overview & Capability Scope\n**Google Sheets Financial Model, KPI Tracker & Formula Playbook** (`skill-google-sheets`) is a verified, portable `SKILL.md` playbook curated with **@levelsio** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nBuild structured Google Sheets ledgers and KPI dashboards programmatically: batch value updates, `INDEX/MATCH` & `QUERY` formulas, conditional formatting, and protected ranges.\n\n### Required Stack & Tooling\n- **Primary Stack:** Google Sheets API, Financial Modeling, KPI Dashboards, Batch Updates\n- **Category:** `finance-deals`\n- **Source & Reference:** [https://developers.google.com/sheets/api](https://developers.google.com/sheets/api)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-google-sheets\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-google-sheets/SKILL.md` or `~/.claude/skills/skill-google-sheets/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx lightstate inspect google-sheets-mcp\n```\n", "category": "finance-deals", "subcategories": [ "finance-deals", "skills" ], "tags": [ "google-sheets-api", "financial-modeling", "kpi-dashboards", "batch-updates", "skill-md", "finance-deals" ], "source_url": "https://developers.google.com/sheets/api", "repo_url": "https://developers.google.com/sheets/api", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-google-sheets", "skill_markdown": "---\nname: google-sheets\ndescription: Build structured Google Sheets ledgers and KPI dashboards programmatically: batch value updates, `INDEX/MATCH` & `QUERY` formulas, conditional formatting, and protected ranges. Use when working with Google Sheets API, Financial Modeling, KPI Dashboards, Batch Updates.\ntriggers:\n - google sheets\n - google sheets api\n - financial modeling\n---\n\n# Google Sheets Financial Model, KPI Tracker & Formula Playbook\n\n## 1. When to Use This Skill\nLoad `skill-google-sheets` whenever an engineering or operations task involves **Google Sheets API, Financial Modeling, KPI Dashboards, Batch Updates**. Specifically:\n- Build structured Google Sheets ledgers and KPI dashboards programmatically: batch value updates, `INDEX/MATCH` & `QUERY` formulas, conditional formatting, and protected ranges.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Google Sheets API, Financial Modeling, KPI Dashboards, Batch Updates\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://developers.google.com/sheets/api`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Google Sheets API**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx lightstate inspect google-sheets-mcp\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Google Sheets Financial Model, KPI Tracker & Formula Playbook (skill-google-sheets) playbook. First inspect our current configuration and codebase for Google Sheets API, Financial Modeling, KPI Dashboards, Batch Updates. Then propose a concrete, step-by-step implementation plan following the skill-google-sheets guardrails, run the non-destructive verification commands (npx lightstate inspect google-sheets-mcp), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Google Sheets API", "Financial Modeling", "KPI Dashboards" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "levelsio", "creator_name": "Pieter Levels (@levelsio)", "creator_url": "https://x.com/levelsio", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 90, "installs_count": 260, "agent_pulls_count": 610, "gravity_score": 2180.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_e35f0e41097a", "slug": "skill-google-drive", "asset_type": "skill", "name": "Google Drive Workspace Taxonomy, Permissions & Export Skill", "tagline": "Organize team Google Drive hierarchies, audit external sharing permissions, convert Docs/Sheets to clean PDFs or Markdown, and index folder metadata withou", "description_md": "### Overview & Capability Scope\n**Google Drive Workspace Taxonomy, Permissions & Export Skill** (`skill-google-drive`) is a verified, portable `SKILL.md` playbook curated with **@petergyang** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nOrganize team Google Drive hierarchies, audit external sharing permissions, convert Docs/Sheets to clean PDFs or Markdown, and index folder metadata without data sprawl.\n\n### Required Stack & Tooling\n- **Primary Stack:** Google Drive API, Document Management, Access Audit, PDF Export\n- **Category:** `operations`\n- **Source & Reference:** [https://developers.google.com/drive/api](https://developers.google.com/drive/api)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-google-drive\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-google-drive/SKILL.md` or `~/.claude/skills/skill-google-drive/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx lightstate inspect google-drive-mcp\n```\n", "category": "operations", "subcategories": [ "operations", "skills" ], "tags": [ "google-drive-api", "document-management", "access-audit", "pdf-export", "skill-md", "operations" ], "source_url": "https://developers.google.com/drive/api", "repo_url": "https://developers.google.com/drive/api", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-google-drive", "skill_markdown": "---\nname: google-drive\ndescription: Organize team Google Drive hierarchies, audit external sharing permissions, convert Docs/Sheets to clean PDFs or Markdown, and index folder metadata without data sprawl. Use when working with Google Drive API, Document Management, Access Audit, PDF Export.\ntriggers:\n - google drive\n - google drive api\n - document management\n---\n\n# Google Drive Workspace Taxonomy, Permissions & Export Skill\n\n## 1. When to Use This Skill\nLoad `skill-google-drive` whenever an engineering or operations task involves **Google Drive API, Document Management, Access Audit, PDF Export**. Specifically:\n- Organize team Google Drive hierarchies, audit external sharing permissions, convert Docs/Sheets to clean PDFs or Markdown, and index folder metadata without data sprawl.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Google Drive API, Document Management, Access Audit, PDF Export\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://developers.google.com/drive/api`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Google Drive API**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx lightstate inspect google-drive-mcp\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Google Drive Workspace Taxonomy, Permissions & Export Skill (skill-google-drive) playbook. First inspect our current configuration and codebase for Google Drive API, Document Management, Access Audit, PDF Export. Then propose a concrete, step-by-step implementation plan following the skill-google-drive guardrails, run the non-destructive verification commands (npx lightstate inspect google-drive-mcp), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Google Drive API", "Document Management", "Access Audit" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "petergyang", "creator_name": "Peter Yang (@petergyang)", "creator_url": "https://x.com/petergyang", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 89, "installs_count": 256, "agent_pulls_count": 603, "gravity_score": 2151.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_f38806d39cad", "slug": "skill-gmail-triage", "asset_type": "skill", "name": "Executive Gmail Priority Triage & Ask-First Draft Workflow Skill", "tagline": "Sweep high-volume email threads, classify messages into Action Required / Waiting / FYI, extract calendar commitments, and stage draft replies with strict ", "description_md": "### Overview & Capability Scope\n**Executive Gmail Priority Triage & Ask-First Draft Workflow Skill** (`skill-gmail-triage`) is a verified, portable `SKILL.md` playbook curated with **@petergyang** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nSweep high-volume email threads, classify messages into Action Required / Waiting / FYI, extract calendar commitments, and stage draft replies with strict human-in-the-loop approval.\n\n### Required Stack & Tooling\n- **Primary Stack:** Gmail API, Inbox Zero, Executive Triage, Ask-First Drafts\n- **Category:** `chief-of-staff`\n- **Source & Reference:** [https://developers.google.com/gmail/api](https://developers.google.com/gmail/api)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-gmail-triage\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-gmail-triage/SKILL.md` or `~/.claude/skills/skill-gmail-triage/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx lightstate inspect gmail-mcp\n```\n", "category": "chief-of-staff", "subcategories": [ "chief-of-staff", "skills" ], "tags": [ "gmail-api", "inbox-zero", "executive-triage", "ask-first-drafts", "skill-md", "chief-of-staff" ], "source_url": "https://developers.google.com/gmail/api", "repo_url": "https://developers.google.com/gmail/api", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-gmail-triage", "skill_markdown": "---\nname: gmail-triage\ndescription: Sweep high-volume email threads, classify messages into Action Required / Waiting / FYI, extract calendar commitments, and stage draft replies with strict human-in-the-loop approval. Use when working with Gmail API, Inbox Zero, Executive Triage, Ask-First Drafts.\ntriggers:\n - gmail triage\n - gmail api\n - inbox zero\n---\n\n# Executive Gmail Priority Triage & Ask-First Draft Workflow Skill\n\n## 1. When to Use This Skill\nLoad `skill-gmail-triage` whenever an engineering or operations task involves **Gmail API, Inbox Zero, Executive Triage, Ask-First Drafts**. Specifically:\n- Sweep high-volume email threads, classify messages into Action Required / Waiting / FYI, extract calendar commitments, and stage draft replies with strict human-in-the-loop approval.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Gmail API, Inbox Zero, Executive Triage, Ask-First Drafts\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://developers.google.com/gmail/api`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Gmail API**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx lightstate inspect gmail-mcp\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Executive Gmail Priority Triage & Ask-First Draft Workflow Skill (skill-gmail-triage) playbook. First inspect our current configuration and codebase for Gmail API, Inbox Zero, Executive Triage, Ask-First Drafts. Then propose a concrete, step-by-step implementation plan following the skill-gmail-triage guardrails, run the non-destructive verification commands (npx lightstate inspect gmail-mcp), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Gmail API", "Inbox Zero", "Executive Triage" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "petergyang", "creator_name": "Peter Yang (@petergyang)", "creator_url": "https://x.com/petergyang", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 88, "installs_count": 252, "agent_pulls_count": 596, "gravity_score": 2122.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_6c3357b7cca8", "slug": "skill-shopify-admin", "asset_type": "skill", "name": "Shopify Admin GraphQL API, Catalog Sync & Order Ops Playbook", "tagline": "Query and mutate Shopify stores using the Admin GraphQL API (`2026-04`): bulk product/variant inventory sync, fulfillment tracking, metaobject schemas, and", "description_md": "### Overview & Capability Scope\n**Shopify Admin GraphQL API, Catalog Sync & Order Ops Playbook** (`skill-shopify-admin`) is a verified, portable `SKILL.md` playbook curated with **@levelsio** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nQuery and mutate Shopify stores using the Admin GraphQL API (`2026-04`): bulk product/variant inventory sync, fulfillment tracking, metaobject schemas, and ShopifyQL analytics.\n\n### Required Stack & Tooling\n- **Primary Stack:** Shopify Admin GraphQL, Shopify CLI, ShopifyQL, Inventory Sync\n- **Category:** `sales-gtm`\n- **Source & Reference:** [https://shopify.dev/docs/api/admin-graphql](https://shopify.dev/docs/api/admin-graphql)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-shopify-admin\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-shopify-admin/SKILL.md` or `~/.claude/skills/skill-shopify-admin/SKILL.md`.\n\n### Verification Commands\n```bash\nshopify app dev\nshopify app deploy\n```\n", "category": "sales-gtm", "subcategories": [ "sales-gtm", "skills" ], "tags": [ "shopify-admin-graphql", "shopify-cli", "shopifyql", "inventory-sync", "skill-md", "sales-gtm" ], "source_url": "https://shopify.dev/docs/api/admin-graphql", "repo_url": "https://shopify.dev/docs/api/admin-graphql", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-shopify-admin", "skill_markdown": "---\nname: shopify-admin\ndescription: Query and mutate Shopify stores using the Admin GraphQL API (`2026-04`): bulk product/variant inventory sync, fulfillment tracking, metaobject schemas, and ShopifyQL analytics. Use when working with Shopify Admin GraphQL, Shopify CLI, ShopifyQL, Inventory Sync.\ntriggers:\n - shopify admin\n - shopify admin graphql\n - shopify cli\n---\n\n# Shopify Admin GraphQL API, Catalog Sync & Order Ops Playbook\n\n## 1. When to Use This Skill\nLoad `skill-shopify-admin` whenever an engineering or operations task involves **Shopify Admin GraphQL, Shopify CLI, ShopifyQL, Inventory Sync**. Specifically:\n- Query and mutate Shopify stores using the Admin GraphQL API (`2026-04`): bulk product/variant inventory sync, fulfillment tracking, metaobject schemas, and ShopifyQL analytics.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Shopify Admin GraphQL, Shopify CLI, ShopifyQL, Inventory Sync\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://shopify.dev/docs/api/admin-graphql`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Shopify Admin GraphQL**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nshopify app dev\nshopify app deploy\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Shopify Admin GraphQL API, Catalog Sync & Order Ops Playbook (skill-shopify-admin) playbook. First inspect our current configuration and codebase for Shopify Admin GraphQL, Shopify CLI, ShopifyQL, Inventory Sync. Then propose a concrete, step-by-step implementation plan following the skill-shopify-admin guardrails, run the non-destructive verification commands (shopify app dev), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Shopify Admin GraphQL", "Shopify CLI", "ShopifyQL" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "levelsio", "creator_name": "Pieter Levels (@levelsio)", "creator_url": "https://x.com/levelsio", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 87, "installs_count": 248, "agent_pulls_count": 589, "gravity_score": 2093.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_8e12debe70a1", "slug": "skill-shopify-functions", "asset_type": "skill", "name": "Shopify Functions Rust/Wasm Discount, Cart & Checkout Skill", "tagline": "Build deterministic sub-5ms backend customizations on Shopify using Shopify Functions: tiered volume discounts, cart validation rules, delivery customizati", "description_md": "### Overview & Capability Scope\n**Shopify Functions Rust/Wasm Discount, Cart & Checkout Skill** (`skill-shopify-functions`) is a verified, portable `SKILL.md` playbook curated with **@levelsio** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nBuild deterministic sub-5ms backend customizations on Shopify using Shopify Functions: tiered volume discounts, cart validation rules, delivery customization, and payment routing.\n\n### Required Stack & Tooling\n- **Primary Stack:** Shopify Functions, WebAssembly, Checkout Extensibility, Discounts API\n- **Category:** `sales-gtm`\n- **Source & Reference:** [https://shopify.dev/docs/api/functions](https://shopify.dev/docs/api/functions)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-shopify-functions\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-shopify-functions/SKILL.md` or `~/.claude/skills/skill-shopify-functions/SKILL.md`.\n\n### Verification Commands\n```bash\nshopify app generate extension --template discount_function\n```\n", "category": "sales-gtm", "subcategories": [ "sales-gtm", "skills" ], "tags": [ "shopify-functions", "webassembly", "checkout-extensibility", "discounts-api", "skill-md", "sales-gtm" ], "source_url": "https://shopify.dev/docs/api/functions", "repo_url": "https://shopify.dev/docs/api/functions", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-shopify-functions", "skill_markdown": "---\nname: shopify-functions\ndescription: Build deterministic sub-5ms backend customizations on Shopify using Shopify Functions: tiered volume discounts, cart validation rules, delivery customization, and payment routing. Use when working with Shopify Functions, WebAssembly, Checkout Extensibility, Discounts API.\ntriggers:\n - shopify functions\n - shopify functions\n - webassembly\n---\n\n# Shopify Functions Rust/Wasm Discount, Cart & Checkout Skill\n\n## 1. When to Use This Skill\nLoad `skill-shopify-functions` whenever an engineering or operations task involves **Shopify Functions, WebAssembly, Checkout Extensibility, Discounts API**. Specifically:\n- Build deterministic sub-5ms backend customizations on Shopify using Shopify Functions: tiered volume discounts, cart validation rules, delivery customization, and payment routing.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Shopify Functions, WebAssembly, Checkout Extensibility, Discounts API\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://shopify.dev/docs/api/functions`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Shopify Functions**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nshopify app generate extension --template discount_function\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Shopify Functions Rust/Wasm Discount, Cart & Checkout Skill (skill-shopify-functions) playbook. First inspect our current configuration and codebase for Shopify Functions, WebAssembly, Checkout Extensibility, Discounts API. Then propose a concrete, step-by-step implementation plan following the skill-shopify-functions guardrails, run the non-destructive verification commands (shopify app generate extension --template discount_function), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Shopify Functions", "WebAssembly", "Checkout Extensibility" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "levelsio", "creator_name": "Pieter Levels (@levelsio)", "creator_url": "https://x.com/levelsio", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 86, "installs_count": 244, "agent_pulls_count": 582, "gravity_score": 2064.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_6a1da79bdc96", "slug": "skill-attio-crm-sync", "asset_type": "skill", "name": "Attio CRM Object Enrichment, Deal Stage & Speed-to-Lead Skill", "tagline": "Sync inbound leads, product signups, and meeting notes into Attio CRM: deduplicate company domains and person records, update deal pipeline stages, and tri", "description_md": "### Overview & Capability Scope\n**Attio CRM Object Enrichment, Deal Stage & Speed-to-Lead Skill** (`skill-attio-crm-sync`) is a verified, portable `SKILL.md` playbook curated with **@ericosiu** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nSync inbound leads, product signups, and meeting notes into Attio CRM: deduplicate company domains and person records, update deal pipeline stages, and trigger founder follow-ups.\n\n### Required Stack & Tooling\n- **Primary Stack:** Attio CRM, Deal Pipelines, Lead Enrichment, RevOps\n- **Category:** `sales-gtm`\n- **Source & Reference:** [https://docs.attio.com](https://docs.attio.com)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-attio-crm-sync\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-attio-crm-sync/SKILL.md` or `~/.claude/skills/skill-attio-crm-sync/SKILL.md`.\n\n### Verification Commands\n```bash\ncurl -s https://api.attio.com/v2/objects -H 'Authorization: Bearer $ATTIO_API_KEY'\n```\n", "category": "sales-gtm", "subcategories": [ "sales-gtm", "skills" ], "tags": [ "attio-crm", "deal-pipelines", "lead-enrichment", "revops", "skill-md", "sales-gtm" ], "source_url": "https://docs.attio.com", "repo_url": "https://docs.attio.com", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-attio-crm-sync", "skill_markdown": "---\nname: attio-crm-sync\ndescription: Sync inbound leads, product signups, and meeting notes into Attio CRM: deduplicate company domains and person records, update deal pipeline stages, and trigger founder follow-ups. Use when working with Attio CRM, Deal Pipelines, Lead Enrichment, RevOps.\ntriggers:\n - attio crm sync\n - attio crm\n - deal pipelines\n---\n\n# Attio CRM Object Enrichment, Deal Stage & Speed-to-Lead Skill\n\n## 1. When to Use This Skill\nLoad `skill-attio-crm-sync` whenever an engineering or operations task involves **Attio CRM, Deal Pipelines, Lead Enrichment, RevOps**. Specifically:\n- Sync inbound leads, product signups, and meeting notes into Attio CRM: deduplicate company domains and person records, update deal pipeline stages, and trigger founder follow-ups.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Attio CRM, Deal Pipelines, Lead Enrichment, RevOps\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://docs.attio.com`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Attio CRM**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\ncurl -s https://api.attio.com/v2/objects -H 'Authorization: Bearer $ATTIO_API_KEY'\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Attio CRM Object Enrichment, Deal Stage & Speed-to-Lead Skill (skill-attio-crm-sync) playbook. First inspect our current configuration and codebase for Attio CRM, Deal Pipelines, Lead Enrichment, RevOps. Then propose a concrete, step-by-step implementation plan following the skill-attio-crm-sync guardrails, run the non-destructive verification commands (curl -s https://api.attio.com/v2/objects -H 'Authorization: Bearer $ATTIO_API_KEY'), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Attio CRM", "Deal Pipelines", "Lead Enrichment" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "ericosiu", "creator_name": "Eric Siu (@ericosiu)", "creator_url": "https://x.com/ericosiu", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 85, "installs_count": 240, "agent_pulls_count": 575, "gravity_score": 2035.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_87b761f86f8d", "slug": "skill-nextjs-app-router", "asset_type": "skill", "name": "Next.js 15+ App Router, Server Actions & Streaming RSC Skill", "tagline": "Architect fast, type-safe Next.js App Router applications: React Server Components (RSC), Suspense streaming boundaries, validated Server Actions, and fine", "description_md": "### Overview & Capability Scope\n**Next.js 15+ App Router, Server Actions & Streaming RSC Skill** (`skill-nextjs-app-router`) is a verified, portable `SKILL.md` playbook curated with **@rauchg** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nArchitect fast, type-safe Next.js App Router applications: React Server Components (RSC), Suspense streaming boundaries, validated Server Actions, and fine-grained cache invalidation.\n\n### Required Stack & Tooling\n- **Primary Stack:** Next.js App Router, React Server Components, Server Actions, Vercel\n- **Category:** `engineering`\n- **Source & Reference:** [https://nextjs.org/docs](https://nextjs.org/docs)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-nextjs-app-router\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-nextjs-app-router/SKILL.md` or `~/.claude/skills/skill-nextjs-app-router/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx next build\nnpx next typegen\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "next.js-app-router", "react-server-components", "server-actions", "vercel", "skill-md", "engineering" ], "source_url": "https://nextjs.org/docs", "repo_url": "https://nextjs.org/docs", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-nextjs-app-router", "skill_markdown": "---\nname: nextjs-app-router\ndescription: Architect fast, type-safe Next.js App Router applications: React Server Components (RSC), Suspense streaming boundaries, validated Server Actions, and fine-grained cache invalidation. Use when working with Next.js App Router, React Server Components, Server Actions, Vercel.\ntriggers:\n - nextjs app router\n - next.js app router\n - react server components\n---\n\n# Next.js 15+ App Router, Server Actions & Streaming RSC Skill\n\n## 1. When to Use This Skill\nLoad `skill-nextjs-app-router` whenever an engineering or operations task involves **Next.js App Router, React Server Components, Server Actions, Vercel**. Specifically:\n- Architect fast, type-safe Next.js App Router applications: React Server Components (RSC), Suspense streaming boundaries, validated Server Actions, and fine-grained cache invalidation.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Next.js App Router, React Server Components, Server Actions, Vercel\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://nextjs.org/docs`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Next.js App Router**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx next build\nnpx next typegen\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Next.js 15+ App Router, Server Actions & Streaming RSC Skill (skill-nextjs-app-router) playbook. First inspect our current configuration and codebase for Next.js App Router, React Server Components, Server Actions, Vercel. Then propose a concrete, step-by-step implementation plan following the skill-nextjs-app-router guardrails, run the non-destructive verification commands (npx next build), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Next.js App Router", "React Server Components", "Server Actions" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "rauchg", "creator_name": "Guillermo Rauch (@rauchg)", "creator_url": "https://x.com/rauchg", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 84, "installs_count": 236, "agent_pulls_count": 568, "gravity_score": 2006.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_e455cd748646", "slug": "skill-anthropic-eval-harness", "asset_type": "skill", "name": "Agentic Tool-Use Evaluation Harness & Prompt Grading Playbook", "tagline": "Design reproducible evaluation suites for tool-calling AI agents: golden test cases, deterministic rubric graders, latency/token cost tracking, and regress", "description_md": "### Overview & Capability Scope\n**Agentic Tool-Use Evaluation Harness & Prompt Grading Playbook** (`skill-anthropic-eval-harness`) is a verified, portable `SKILL.md` playbook curated with **@alexalbert__** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nDesign reproducible evaluation suites for tool-calling AI agents: golden test cases, deterministic rubric graders, latency/token cost tracking, and regression diffs across prompt edits.\n\n### Required Stack & Tooling\n- **Primary Stack:** LLM Evals, Tool-Use Benchmarks, System Prompts, Rubric Grading\n- **Category:** `research`\n- **Source & Reference:** [https://github.com/anthropics/courses](https://github.com/anthropics/courses)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-anthropic-eval-harness\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-anthropic-eval-harness/SKILL.md` or `~/.claude/skills/skill-anthropic-eval-harness/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx promptfoo@latest eval\nnpx promptfoo@latest view\n```\n", "category": "research", "subcategories": [ "research", "skills" ], "tags": [ "llm-evals", "tool-use-benchmarks", "system-prompts", "rubric-grading", "skill-md", "research" ], "source_url": "https://github.com/anthropics/courses", "repo_url": "https://github.com/anthropics/courses", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-anthropic-eval-harness", "skill_markdown": "---\nname: anthropic-eval-harness\ndescription: Design reproducible evaluation suites for tool-calling AI agents: golden test cases, deterministic rubric graders, latency/token cost tracking, and regression diffs across prompt edits. Use when working with LLM Evals, Tool-Use Benchmarks, System Prompts, Rubric Grading.\ntriggers:\n - anthropic eval harness\n - llm evals\n - tool-use benchmarks\n---\n\n# Agentic Tool-Use Evaluation Harness & Prompt Grading Playbook\n\n## 1. When to Use This Skill\nLoad `skill-anthropic-eval-harness` whenever an engineering or operations task involves **LLM Evals, Tool-Use Benchmarks, System Prompts, Rubric Grading**. Specifically:\n- Design reproducible evaluation suites for tool-calling AI agents: golden test cases, deterministic rubric graders, latency/token cost tracking, and regression diffs across prompt edits.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** LLM Evals, Tool-Use Benchmarks, System Prompts, Rubric Grading\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/anthropics/courses`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **LLM Evals**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx promptfoo@latest eval\nnpx promptfoo@latest view\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Agentic Tool-Use Evaluation Harness & Prompt Grading Playbook (skill-anthropic-eval-harness) playbook. First inspect our current configuration and codebase for LLM Evals, Tool-Use Benchmarks, System Prompts, Rubric Grading. Then propose a concrete, step-by-step implementation plan following the skill-anthropic-eval-harness guardrails, run the non-destructive verification commands (npx promptfoo@latest eval), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "LLM Evals", "Tool-Use Benchmarks", "System Prompts" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "alexalbert__", "creator_name": "Alex Albert (@alexalbert__)", "creator_url": "https://x.com/alexalbert__", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 94, "installs_count": 228, "agent_pulls_count": 554, "gravity_score": 1978.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_fe042aa60194", "slug": "skill-shadcn-design-system", "asset_type": "skill", "name": "shadcn/ui, Radix Primitives & Dark-Mode Design Token Playbook", "tagline": "Build accessible, keyboard-navigable interfaces with shadcn/ui and Tailwind CSS: semantic CSS variables, WCAG AA contrast verification, focus-ring hygiene,", "description_md": "### Overview & Capability Scope\n**shadcn/ui, Radix Primitives & Dark-Mode Design Token Playbook** (`skill-shadcn-design-system`) is a verified, portable `SKILL.md` playbook curated with **@shadcn** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nBuild accessible, keyboard-navigable interfaces with shadcn/ui and Tailwind CSS: semantic CSS variables, WCAG AA contrast verification, focus-ring hygiene, and zero-CLS dialogs.\n\n### Required Stack & Tooling\n- **Primary Stack:** shadcn/ui, Radix UI, Tailwind CSS, Design Tokens\n- **Category:** `creative-video`\n- **Source & Reference:** [https://ui.shadcn.com/docs](https://ui.shadcn.com/docs)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-shadcn-design-system\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-shadcn-design-system/SKILL.md` or `~/.claude/skills/skill-shadcn-design-system/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx shadcn@latest add button dialog table\nnpx shadcn@latest diff\n```\n", "category": "creative-video", "subcategories": [ "creative-video", "skills" ], "tags": [ "shadcn/ui", "radix-ui", "tailwind-css", "design-tokens", "skill-md", "creative-video" ], "source_url": "https://ui.shadcn.com/docs", "repo_url": "https://ui.shadcn.com/docs", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-shadcn-design-system", "skill_markdown": "---\nname: shadcn-design-system\ndescription: Build accessible, keyboard-navigable interfaces with shadcn/ui and Tailwind CSS: semantic CSS variables, WCAG AA contrast verification, focus-ring hygiene, and zero-CLS dialogs. Use when working with shadcn/ui, Radix UI, Tailwind CSS, Design Tokens.\ntriggers:\n - shadcn design system\n - shadcn/ui\n - radix ui\n---\n\n# shadcn/ui, Radix Primitives & Dark-Mode Design Token Playbook\n\n## 1. When to Use This Skill\nLoad `skill-shadcn-design-system` whenever an engineering or operations task involves **shadcn/ui, Radix UI, Tailwind CSS, Design Tokens**. Specifically:\n- Build accessible, keyboard-navigable interfaces with shadcn/ui and Tailwind CSS: semantic CSS variables, WCAG AA contrast verification, focus-ring hygiene, and zero-CLS dialogs.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** shadcn/ui, Radix UI, Tailwind CSS, Design Tokens\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://ui.shadcn.com/docs`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **shadcn/ui**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx shadcn@latest add button dialog table\nnpx shadcn@latest diff\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the shadcn/ui, Radix Primitives & Dark-Mode Design Token Playbook (skill-shadcn-design-system) playbook. First inspect our current configuration and codebase for shadcn/ui, Radix UI, Tailwind CSS, Design Tokens. Then propose a concrete, step-by-step implementation plan following the skill-shadcn-design-system guardrails, run the non-destructive verification commands (npx shadcn@latest add button dialog table), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "shadcn/ui", "Radix UI", "Tailwind CSS" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "shadcn", "creator_name": "shadcn (@shadcn)", "creator_url": "https://x.com/shadcn", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 83, "installs_count": 232, "agent_pulls_count": 561, "gravity_score": 1977.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_5b526b8497eb", "slug": "skill-skill-creator", "asset_type": "skill", "name": "Portable SKILL.md Authoring, Trigger Optimization & QA Playbook", "tagline": "Author, test, and package reusable `SKILL.md` playbooks compatible with Grok Bot, Claude Code, and Cursor: YAML frontmatter trigger phrasing, progressive d", "description_md": "### Overview & Capability Scope\n**Portable SKILL.md Authoring, Trigger Optimization & QA Playbook** (`skill-skill-creator`) is a verified, portable `SKILL.md` playbook curated with **@alexalbert__** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nAuthor, test, and package reusable `SKILL.md` playbooks compatible with Grok Bot, Claude Code, and Cursor: YAML frontmatter trigger phrasing, progressive disclosure, and verification scripts.\n\n### Required Stack & Tooling\n- **Primary Stack:** SKILL.md Spec, Agent Playbooks, Trigger Tuning, Progressive Disclosure\n- **Category:** `engineering`\n- **Source & Reference:** [https://github.com/anthropics/skills](https://github.com/anthropics/skills)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-skill-creator\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-skill-creator/SKILL.md` or `~/.claude/skills/skill-skill-creator/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx lightstate inspect skill-skill-creator\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "skill.md-spec", "agent-playbooks", "trigger-tuning", "progressive-disclosure", "skill-md", "engineering" ], "source_url": "https://github.com/anthropics/skills", "repo_url": "https://github.com/anthropics/skills", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-skill-creator", "skill_markdown": "---\nname: creator\ndescription: Author, test, and package reusable `SKILL.md` playbooks compatible with Grok Bot, Claude Code, and Cursor: YAML frontmatter trigger phrasing, progressive disclosure, and verification scripts. Use when working with SKILL.md Spec, Agent Playbooks, Trigger Tuning, Progressive Disclosure.\ntriggers:\n - creator\n - skill.md spec\n - agent playbooks\n---\n\n# Portable SKILL.md Authoring, Trigger Optimization & QA Playbook\n\n## 1. When to Use This Skill\nLoad `skill-skill-creator` whenever an engineering or operations task involves **SKILL.md Spec, Agent Playbooks, Trigger Tuning, Progressive Disclosure**. Specifically:\n- Author, test, and package reusable `SKILL.md` playbooks compatible with Grok Bot, Claude Code, and Cursor: YAML frontmatter trigger phrasing, progressive disclosure, and verification scripts.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** SKILL.md Spec, Agent Playbooks, Trigger Tuning, Progressive Disclosure\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://github.com/anthropics/skills`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **SKILL.md Spec**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx lightstate inspect skill-skill-creator\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Portable SKILL.md Authoring, Trigger Optimization & QA Playbook (skill-skill-creator) playbook. First inspect our current configuration and codebase for SKILL.md Spec, Agent Playbooks, Trigger Tuning, Progressive Disclosure. Then propose a concrete, step-by-step implementation plan following the skill-skill-creator guardrails, run the non-destructive verification commands (npx lightstate inspect skill-skill-creator), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "SKILL.md Spec", "Agent Playbooks", "Trigger Tuning" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "alexalbert__", "creator_name": "Alex Albert (@alexalbert__)", "creator_url": "https://x.com/alexalbert__", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 93, "installs_count": 224, "agent_pulls_count": 547, "gravity_score": 1949.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_1f45a111262c", "slug": "skill-technical-seo-aeo-audit", "asset_type": "skill", "name": "Technical SEO, Schema.org JSON-LD & AI Answer Engine (AEO) Skill", "tagline": "Audit and implement programmatic SEO & AI Answer Engine Optimization (AEO): canonical tags, XML sitemaps, `llms.txt` / `llms-full.txt`, `SoftwareApplicatio", "description_md": "### Overview & Capability Scope\n**Technical SEO, Schema.org JSON-LD & AI Answer Engine (AEO) Skill** (`skill-technical-seo-aeo-audit`) is a verified, portable `SKILL.md` playbook curated with **@ericosiu** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nAudit and implement programmatic SEO & AI Answer Engine Optimization (AEO): canonical tags, XML sitemaps, `llms.txt` / `llms-full.txt`, `SoftwareApplication` & `FAQPage` JSON-LD.\n\n### Required Stack & Tooling\n- **Primary Stack:** Technical SEO, AEO / llms.txt, JSON-LD Schema, Sitemap XML\n- **Category:** `marketing-seo`\n- **Source & Reference:** [https://schema.org/docs/schemas.html](https://schema.org/docs/schemas.html)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-technical-seo-aeo-audit\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-technical-seo-aeo-audit/SKILL.md` or `~/.claude/skills/skill-technical-seo-aeo-audit/SKILL.md`.\n\n### Verification Commands\n```bash\ncurl -sI https://example.com/llms.txt\ncurl -sI https://example.com/sitemap.xml\n```\n", "category": "marketing-seo", "subcategories": [ "marketing-seo", "skills" ], "tags": [ "technical-seo", "aeo-/-llms.txt", "json-ld-schema", "sitemap-xml", "skill-md", "marketing-seo" ], "source_url": "https://schema.org/docs/schemas.html", "repo_url": "https://schema.org/docs/schemas.html", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-technical-seo-aeo-audit", "skill_markdown": "---\nname: technical-seo-aeo-audit\ndescription: Audit and implement programmatic SEO & AI Answer Engine Optimization (AEO): canonical tags, XML sitemaps, `llms.txt` / `llms-full.txt`, `SoftwareApplication` & `FAQPage` JSON-LD. Use when working with Technical SEO, AEO / llms.txt, JSON-LD Schema, Sitemap XML.\ntriggers:\n - technical seo aeo audit\n - technical seo\n - aeo / llms.txt\n---\n\n# Technical SEO, Schema.org JSON-LD & AI Answer Engine (AEO) Skill\n\n## 1. When to Use This Skill\nLoad `skill-technical-seo-aeo-audit` whenever an engineering or operations task involves **Technical SEO, AEO / llms.txt, JSON-LD Schema, Sitemap XML**. Specifically:\n- Audit and implement programmatic SEO & AI Answer Engine Optimization (AEO): canonical tags, XML sitemaps, `llms.txt` / `llms-full.txt`, `SoftwareApplication` & `FAQPage` JSON-LD.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Technical SEO, AEO / llms.txt, JSON-LD Schema, Sitemap XML\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://schema.org/docs/schemas.html`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Technical SEO**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\ncurl -sI https://example.com/llms.txt\ncurl -sI https://example.com/sitemap.xml\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Technical SEO, Schema.org JSON-LD & AI Answer Engine (AEO) Skill (skill-technical-seo-aeo-audit) playbook. First inspect our current configuration and codebase for Technical SEO, AEO / llms.txt, JSON-LD Schema, Sitemap XML. Then propose a concrete, step-by-step implementation plan following the skill-technical-seo-aeo-audit guardrails, run the non-destructive verification commands (curl -sI https://example.com/llms.txt), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Technical SEO", "AEO / llms.txt", "JSON-LD Schema" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "ericosiu", "creator_name": "Eric Siu (@ericosiu)", "creator_url": "https://x.com/ericosiu", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 92, "installs_count": 220, "agent_pulls_count": 540, "gravity_score": 1920.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_52816b13aca1", "slug": "skill-monorepo-turborepo", "asset_type": "skill", "name": "Turborepo & pnpm Workspace Remote Caching & Task Graph Skill", "tagline": "Configure high-speed TypeScript monorepos with pnpm workspaces and Turborepo: deterministic task pipelines (`build`, `lint`, `test`), input hashing, and re", "description_md": "### Overview & Capability Scope\n**Turborepo & pnpm Workspace Remote Caching & Task Graph Skill** (`skill-monorepo-turborepo`) is a verified, portable `SKILL.md` playbook curated with **@mattpocockuk** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nConfigure high-speed TypeScript monorepos with pnpm workspaces and Turborepo: deterministic task pipelines (`build`, `lint`, `test`), input hashing, and remote artifact caching.\n\n### Required Stack & Tooling\n- **Primary Stack:** Turborepo, pnpm Workspaces, Monorepo Architecture, Remote Caching\n- **Category:** `engineering`\n- **Source & Reference:** [https://turbo.build/repo/docs](https://turbo.build/repo/docs)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-monorepo-turborepo\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-monorepo-turborepo/SKILL.md` or `~/.claude/skills/skill-monorepo-turborepo/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx turbo run build test --summarize\nnpx turbo run lint --affected\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "turborepo", "pnpm-workspaces", "monorepo-architecture", "remote-caching", "skill-md", "engineering" ], "source_url": "https://turbo.build/repo/docs", "repo_url": "https://turbo.build/repo/docs", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-monorepo-turborepo", "skill_markdown": "---\nname: monorepo-turborepo\ndescription: Configure high-speed TypeScript monorepos with pnpm workspaces and Turborepo: deterministic task pipelines (`build`, `lint`, `test`), input hashing, and remote artifact caching. Use when working with Turborepo, pnpm Workspaces, Monorepo Architecture, Remote Caching.\ntriggers:\n - monorepo turborepo\n - turborepo\n - pnpm workspaces\n---\n\n# Turborepo & pnpm Workspace Remote Caching & Task Graph Skill\n\n## 1. When to Use This Skill\nLoad `skill-monorepo-turborepo` whenever an engineering or operations task involves **Turborepo, pnpm Workspaces, Monorepo Architecture, Remote Caching**. Specifically:\n- Configure high-speed TypeScript monorepos with pnpm workspaces and Turborepo: deterministic task pipelines (`build`, `lint`, `test`), input hashing, and remote artifact caching.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Turborepo, pnpm Workspaces, Monorepo Architecture, Remote Caching\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://turbo.build/repo/docs`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Turborepo**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx turbo run build test --summarize\nnpx turbo run lint --affected\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Turborepo & pnpm Workspace Remote Caching & Task Graph Skill (skill-monorepo-turborepo) playbook. First inspect our current configuration and codebase for Turborepo, pnpm Workspaces, Monorepo Architecture, Remote Caching. Then propose a concrete, step-by-step implementation plan following the skill-monorepo-turborepo guardrails, run the non-destructive verification commands (npx turbo run build test --summarize), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Turborepo", "pnpm Workspaces", "Monorepo Architecture" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "mattpocockuk", "creator_name": "Matt Pocock (@mattpocockuk)", "creator_url": "https://x.com/mattpocockuk", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 91, "installs_count": 216, "agent_pulls_count": 533, "gravity_score": 1891.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_4a58b049731d", "slug": "skill-sandbox-code-execution", "asset_type": "skill", "name": "Isolated Cloud Sandbox Execution & Untrusted Code Runner Skill", "tagline": "Execute LLM-generated Python, Node.js, and shell scripts safely inside ephemeral microVM sandboxes with strict network egress controls, filesystem quotas, ", "description_md": "### Overview & Capability Scope\n**Isolated Cloud Sandbox Execution & Untrusted Code Runner Skill** (`skill-sandbox-code-execution`) is a verified, portable `SKILL.md` playbook curated with **@amasad** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nExecute LLM-generated Python, Node.js, and shell scripts safely inside ephemeral microVM sandboxes with strict network egress controls, filesystem quotas, and timeout enforcement.\n\n### Required Stack & Tooling\n- **Primary Stack:** E2B Sandbox, Cloudflare Sandbox, MicroVM Isolation, Code Interpreter\n- **Category:** `engineering`\n- **Source & Reference:** [https://e2b.dev/docs](https://e2b.dev/docs)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-sandbox-code-execution\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-sandbox-code-execution/SKILL.md` or `~/.claude/skills/skill-sandbox-code-execution/SKILL.md`.\n\n### Verification Commands\n```bash\nnpm i @e2b/code-interpreter\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "e2b-sandbox", "cloudflare-sandbox", "microvm-isolation", "code-interpreter", "skill-md", "engineering" ], "source_url": "https://e2b.dev/docs", "repo_url": "https://e2b.dev/docs", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-sandbox-code-execution", "skill_markdown": "---\nname: sandbox-code-execution\ndescription: Execute LLM-generated Python, Node.js, and shell scripts safely inside ephemeral microVM sandboxes with strict network egress controls, filesystem quotas, and timeout enforcement. Use when working with E2B Sandbox, Cloudflare Sandbox, MicroVM Isolation, Code Interpreter.\ntriggers:\n - sandbox code execution\n - e2b sandbox\n - cloudflare sandbox\n---\n\n# Isolated Cloud Sandbox Execution & Untrusted Code Runner Skill\n\n## 1. When to Use This Skill\nLoad `skill-sandbox-code-execution` whenever an engineering or operations task involves **E2B Sandbox, Cloudflare Sandbox, MicroVM Isolation, Code Interpreter**. Specifically:\n- Execute LLM-generated Python, Node.js, and shell scripts safely inside ephemeral microVM sandboxes with strict network egress controls, filesystem quotas, and timeout enforcement.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** E2B Sandbox, Cloudflare Sandbox, MicroVM Isolation, Code Interpreter\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://e2b.dev/docs`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **E2B Sandbox**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpm i @e2b/code-interpreter\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Isolated Cloud Sandbox Execution & Untrusted Code Runner Skill (skill-sandbox-code-execution) playbook. First inspect our current configuration and codebase for E2B Sandbox, Cloudflare Sandbox, MicroVM Isolation, Code Interpreter. Then propose a concrete, step-by-step implementation plan following the skill-sandbox-code-execution guardrails, run the non-destructive verification commands (npm i @e2b/code-interpreter), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "E2B Sandbox", "Cloudflare Sandbox", "MicroVM Isolation" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "amasad", "creator_name": "Amjad Masad (@amasad)", "creator_url": "https://x.com/amasad", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 90, "installs_count": 212, "agent_pulls_count": 526, "gravity_score": 1862.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_1bced52d87ae", "slug": "skill-rag-hybrid-search", "asset_type": "skill", "name": "Hybrid Vector + BM25 Full-Text Search & Reranking RAG Skill", "tagline": "Build high-precision Retrieval-Augmented Generation (RAG) pipelines combining dense vector embeddings, sparse BM25 keyword matching, Reciprocal Rank Fusion", "description_md": "### Overview & Capability Scope\n**Hybrid Vector + BM25 Full-Text Search & Reranking RAG Skill** (`skill-rag-hybrid-search`) is a verified, portable `SKILL.md` playbook curated with **@swyx** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nBuild high-precision Retrieval-Augmented Generation (RAG) pipelines combining dense vector embeddings, sparse BM25 keyword matching, Reciprocal Rank Fusion (RRF), and cross-encoder reranking.\n\n### Required Stack & Tooling\n- **Primary Stack:** Hybrid Search, Reciprocal Rank Fusion, Vector DB, Citations\n- **Category:** `research`\n- **Source & Reference:** [https://qdrant.tech/documentation/](https://qdrant.tech/documentation/)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-rag-hybrid-search\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-rag-hybrid-search/SKILL.md` or `~/.claude/skills/skill-rag-hybrid-search/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx lightstate inspect qdrant-mcp\n```\n", "category": "research", "subcategories": [ "research", "skills" ], "tags": [ "hybrid-search", "reciprocal-rank-fusion", "vector-db", "citations", "skill-md", "research" ], "source_url": "https://qdrant.tech/documentation/", "repo_url": "https://qdrant.tech/documentation/", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-rag-hybrid-search", "skill_markdown": "---\nname: rag-hybrid-search\ndescription: Build high-precision Retrieval-Augmented Generation (RAG) pipelines combining dense vector embeddings, sparse BM25 keyword matching, Reciprocal Rank Fusion (RRF), and cross-encoder reranking. Use when working with Hybrid Search, Reciprocal Rank Fusion, Vector DB, Citations.\ntriggers:\n - rag hybrid search\n - hybrid search\n - reciprocal rank fusion\n---\n\n# Hybrid Vector + BM25 Full-Text Search & Reranking RAG Skill\n\n## 1. When to Use This Skill\nLoad `skill-rag-hybrid-search` whenever an engineering or operations task involves **Hybrid Search, Reciprocal Rank Fusion, Vector DB, Citations**. Specifically:\n- Build high-precision Retrieval-Augmented Generation (RAG) pipelines combining dense vector embeddings, sparse BM25 keyword matching, Reciprocal Rank Fusion (RRF), and cross-encoder reranking.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Hybrid Search, Reciprocal Rank Fusion, Vector DB, Citations\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://qdrant.tech/documentation/`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Hybrid Search**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx lightstate inspect qdrant-mcp\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Hybrid Vector + BM25 Full-Text Search & Reranking RAG Skill (skill-rag-hybrid-search) playbook. First inspect our current configuration and codebase for Hybrid Search, Reciprocal Rank Fusion, Vector DB, Citations. Then propose a concrete, step-by-step implementation plan following the skill-rag-hybrid-search guardrails, run the non-destructive verification commands (npx lightstate inspect qdrant-mcp), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Hybrid Search", "Reciprocal Rank Fusion", "Vector DB" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "swyx", "creator_name": "Shawn Wang (@swyx)", "creator_url": "https://x.com/swyx", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 89, "installs_count": 208, "agent_pulls_count": 519, "gravity_score": 1833.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_8b772b805db6", "slug": "skill-readme-docs-generator", "asset_type": "skill", "name": "Open-Source README, Architecture Diagram & Mintlify Docs Skill", "tagline": "Generate developer-loved documentation from source code: quickstart CLI blocks, Mermaid system architecture diagrams, API parameter tables, and runnable co", "description_md": "### Overview & Capability Scope\n**Open-Source README, Architecture Diagram & Mintlify Docs Skill** (`skill-readme-docs-generator`) is a verified, portable `SKILL.md` playbook curated with **@nutlope** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nGenerate developer-loved documentation from source code: quickstart CLI blocks, Mermaid system architecture diagrams, API parameter tables, and runnable copy-paste examples.\n\n### Required Stack & Tooling\n- **Primary Stack:** Mintlify, Mermaid Diagrams, Developer Docs, OpenAPI\n- **Category:** `engineering`\n- **Source & Reference:** [https://mintlify.com/docs](https://mintlify.com/docs)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-readme-docs-generator\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-readme-docs-generator/SKILL.md` or `~/.claude/skills/skill-readme-docs-generator/SKILL.md`.\n\n### Verification Commands\n```bash\nnpx mintlify dev\nnpx mintlify broken-links\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "mintlify", "mermaid-diagrams", "developer-docs", "openapi", "skill-md", "engineering" ], "source_url": "https://mintlify.com/docs", "repo_url": "https://mintlify.com/docs", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-readme-docs-generator", "skill_markdown": "---\nname: readme-docs-generator\ndescription: Generate developer-loved documentation from source code: quickstart CLI blocks, Mermaid system architecture diagrams, API parameter tables, and runnable copy-paste examples. Use when working with Mintlify, Mermaid Diagrams, Developer Docs, OpenAPI.\ntriggers:\n - readme docs generator\n - mintlify\n - mermaid diagrams\n---\n\n# Open-Source README, Architecture Diagram & Mintlify Docs Skill\n\n## 1. When to Use This Skill\nLoad `skill-readme-docs-generator` whenever an engineering or operations task involves **Mintlify, Mermaid Diagrams, Developer Docs, OpenAPI**. Specifically:\n- Generate developer-loved documentation from source code: quickstart CLI blocks, Mermaid system architecture diagrams, API parameter tables, and runnable copy-paste examples.\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Mintlify, Mermaid Diagrams, Developer Docs, OpenAPI\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://mintlify.com/docs`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Mintlify**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpx mintlify dev\nnpx mintlify broken-links\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Open-Source README, Architecture Diagram & Mintlify Docs Skill (skill-readme-docs-generator) playbook. First inspect our current configuration and codebase for Mintlify, Mermaid Diagrams, Developer Docs, OpenAPI. Then propose a concrete, step-by-step implementation plan following the skill-readme-docs-generator guardrails, run the non-destructive verification commands (npx mintlify dev), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Mintlify", "Mermaid Diagrams", "Developer Docs" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "nutlope", "creator_name": "Hassan El Mghari (@nutlope)", "creator_url": "https://x.com/nutlope", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 88, "installs_count": 204, "agent_pulls_count": 512, "gravity_score": 1804.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" }, { "id": "skl_2d288a329c91", "slug": "skill-vercel-ai-sdk", "asset_type": "skill", "name": "Vercel AI SDK 4.x Core, `streamText` & Multi-Step Tool Calling Skill", "tagline": "Build streaming, tool-calling AI agents in TypeScript using the Vercel AI SDK (`streamText`, `generateObject`, `maxSteps`, and provider-agnostic model rout", "description_md": "### Overview & Capability Scope\n**Vercel AI SDK 4.x Core, `streamText` & Multi-Step Tool Calling Skill** (`skill-vercel-ai-sdk`) is a verified, portable `SKILL.md` playbook curated with **@vercel** for Grok Bot (`x.ai/bot/`), Claude Code (`~/.claude/skills/`), and Cursor (`.cursor/skills/`).\n\nBuild streaming, tool-calling AI agents in TypeScript using the Vercel AI SDK (`streamText`, `generateObject`, `maxSteps`, and provider-agnostic model routing).\n\n### Required Stack & Tooling\n- **Primary Stack:** Vercel AI SDK, TypeScript, Streaming UI, Zod Schemas\n- **Category:** `engineering`\n- **Source & Reference:** [https://sdk.vercel.ai/docs](https://sdk.vercel.ai/docs)\n\n### Quickstart Installation\nInstall this `SKILL.md` playbook directly into your agent workspace:\n```bash\nnpx lightstate install skill-vercel-ai-sdk\n```\nOr copy the complete `SKILL.md` specification below into `.cursor/skills/skill-vercel-ai-sdk/SKILL.md` or `~/.claude/skills/skill-vercel-ai-sdk/SKILL.md`.\n\n### Verification Commands\n```bash\nnpm i ai @ai-sdk/openai @ai-sdk/anthropic\nnpx next build\n```\n", "category": "engineering", "subcategories": [ "engineering", "skills" ], "tags": [ "vercel-ai-sdk", "typescript", "streaming-ui", "zod-schemas", "skill-md", "engineering" ], "source_url": "https://sdk.vercel.ai/docs", "repo_url": "https://sdk.vercel.ai/docs", "bot_share_url": "", "bot_id": "", "deep_link_url": "", "mcp_endpoint_url": "", "mcp_transport": "", "mcp_auth_type": "", "install_command": "npx lightstate install skill-vercel-ai-sdk", "skill_markdown": "---\nname: vercel-ai-sdk\ndescription: Build streaming, tool-calling AI agents in TypeScript using the Vercel AI SDK (`streamText`, `generateObject`, `maxSteps`, and provider-agnostic model routing). Use when working with Vercel AI SDK, TypeScript, Streaming UI, Zod Schemas.\ntriggers:\n - vercel ai sdk\n - vercel ai sdk\n - typescript\n---\n\n# Vercel AI SDK 4.x Core, `streamText` & Multi-Step Tool Calling Skill\n\n## 1. When to Use This Skill\nLoad `skill-vercel-ai-sdk` whenever an engineering or operations task involves **Vercel AI SDK, TypeScript, Streaming UI, Zod Schemas**. Specifically:\n- Build streaming, tool-calling AI agents in TypeScript using the Vercel AI SDK (`streamText`, `generateObject`, `maxSteps`, and provider-agnostic model routing).\n- Designing, reviewing, or debugging production configurations without guesswork or deprecated flags.\n- Running deterministic pre-flight and post-execution verification checks before reporting completion.\n\n## 2. Prerequisites & Required Environment\n- **Target Stack:** Vercel AI SDK, TypeScript, Streaming UI, Zod Schemas\n- **Permissions:** Read-only inspection by default; ask for explicit confirmation before executing destructive mutations or production deployments.\n- **Reference Repository:** `https://sdk.vercel.ai/docs`\n\n## 3. Step-by-Step Execution Playbook\n1. **Context & Configuration Discovery:**\n Inspect existing project manifests (`package.json`, `wrangler.jsonc`, schema files, or environment declarations) before proposing changes. Never overwrite existing bindings or credentials blindly.\n2. **Deterministic Implementation:**\n Apply the canonical patterns for **Vercel AI SDK**:\n - Enforce strict schema/type validation at every external boundary.\n - Include explicit error handling, structured logging, and idempotency keys on all mutating calls.\n - Keep secrets in environment variables or secret stores—never hardcode tokens in source files.\n3. **Verification & Regression Gate:**\n Run the verification commands below and inspect both exit codes and output payloads before marking the task complete.\n\n## 4. Reference Commands & CLI Workflow\n```bash\nnpm i ai @ai-sdk/openai @ai-sdk/anthropic\nnpx next build\n```\n\n## 5. Edge Cases & Anti-Hallucination Guardrails\n- **Zero Unverified Claims:** Never claim a deployment, migration, or API call succeeded without inspecting the command output or HTTP status code.\n- **Rollback Readiness:** Always record the previous state or migration version so changes can be reverted cleanly if downstream health checks fail.\n", "starter_prompt": "Load the Vercel AI SDK 4.x Core, `streamText` & Multi-Step Tool Calling Skill (skill-vercel-ai-sdk) playbook. First inspect our current configuration and codebase for Vercel AI SDK, TypeScript, Streaming UI, Zod Schemas. Then propose a concrete, step-by-step implementation plan following the skill-vercel-ai-sdk guardrails, run the non-destructive verification commands (npm i ai @ai-sdk/openai @ai-sdk/anthropic), and summarize any findings, risks, or required environment variables before making production changes.", "Required_Connectors": [ "Vercel AI SDK", "TypeScript", "Streaming UI" ], "Permissions_Safety_Level": "Read-Only / Ask-First", "creator_handle": "vercel", "creator_name": "Vercel Engineering", "creator_url": "https://x.com/vercel", "verified": true, "claimed": false, "is_sponsored": false, "sponsor_tier": "free", "sponsor_cta_text": "", "sponsor_cta_url": "", "pulse_score": 87, "installs_count": 200, "agent_pulls_count": 505, "gravity_score": 1775.0, "is_featured": false, "created_at": "2026-10-24T20:00:00Z", "updated_at": "2026-10-24T20:00:00Z" } ] }