Context & AI engineering
25,729 rows, by stacks then stars
1,865 of these skill files have been read, by 1,328 distinct authors, and what they tell an agent counted. What Context & AI engineering authors agree on, and what they forbid
stevennitesh/programming-agent-skills/skills/.archive/subagent-workflow Skill
1★ repoUse when the user explicitly asks for subagents, parallel agents, worker agents, independent code review, multi-agent codebase exploration, subagent-ready handoffs, or an approved plan/issue calls for authorized, bounded delegation that is cheaper than local work.
stevennitesh/programming-agent-skills/skills/.archive/worktree-isolation Skill
1★ repoUse when approved parallel implementation across agents or sessions needs separate git branches/worktrees, when a user explicitly asks for isolated worktrees, or when `parallel-disjoint`/`parallel-overlap` execution needs branch, ownership, integration, or cleanup safety before editing.
stevennitesh/programming-agent-skills/skills/custom/handoff Skill
1★ repoCompact the live thread into a resumable handoff for a fresh Codex session or agent thread.
jpab/maieutic-tools/tools/socratic-dev/skills/socratic-dev Skill
1★ repoGuided agentic development loop. Surfaces ambiguities, checks the task is plannable, forces an option choice, and waits for explicit plan approval before any write-capable agent runs.
simonkoehler/event-skills/event-orchestrator Skill
1★ repoCoordinate multi-skill workflows for complex B2B event projects. Analyzes the user's request, determines the optimal skill execution order, and manages context passing between skills. Use this skill when a user's request spans multiple event domains (e.g., "plan a full conference" requires planner + budget + speakers + content + logistics), when the user is unsure which skill to start with, or when outputs from one skill need to feed into another. Acts as the intelligent router and workflow manager for the event-management-skills collection.
krlan/leverbrain/skills/claude-design Skill
no license1★ repoWeb3 skill marketplace for agents and operators
harrylabsj/hermes-skills/skills/token-cost-guard Skill
no license1★ repoMonitor agent token usage and model cost from the active ecosystem's own records, compare the current run with the previous snapshot, and send an alert report when cost growth exceeds a threshold. In OpenClaw it reads OpenClaw session logs; in Hermes agent it reads Hermes logs/state.db instead of OpenClaw. Use when users ask to track token spend, calculate realtime token cost, detect spending spikes, monitor DeepSeek/Kimi/GPT model usage, or set up cost alerts for OpenClaw or Hermes agents.
GustavoGutierrez/engineering-skills/skills/prompt-engineer Skill
1★ repoDesign and engineer System Prompts, prompt templates, and multi-agent orchestration contracts for deterministic, leak-proof AI systems. Use when creating agents, writing skill definitions, designing prompt templates with safe variable injection, structuring I/O contracts, or building multi-agent pipelines.
jsirish/workflow-skills/skills/handoff Skill
no license1★ repoUpdates the master HANDOFF.md with distilled knowledge and creates a timestamped session log. Run ONLY when explicitly invoked by the user via /handoff. Portable no-plugin fallback: if the throughline plugin is active, use throughline-handoff instead.
jsirish/workflow-skills/skills/pr-review Skill
no license1★ repoRequest and manage pull request reviews using pr-agent (local CLI) with optional Claude second-pass. Use when needing to review a PR, check review comments, address feedback, or iterate on fixes. Triggers on "review pr", "request review", "check review comments", "review this", or "re-review".
The same file is in 1 other repository
sainzs/santiagosainz-skills/skills/core/handoff Skill
1★ repoWrite a durable engineering handoff for ongoing agent work: changed files, current goal, evidence, risks, blocked items, and restart prompt. Use before pausing, compacting context, switching tools, ending a long session, or using work-handoff.
rmerk/crewai-headless-flow/.agents/skills/git-guardrails-claude-code Skill
1★ repoSet up Claude Code hooks to block dangerous git commands (push, reset --hard, clean, branch -D, etc.) before they execute. Use when user wants to prevent destructive git operations, add git safety hooks, or block git push/reset in Claude Code.
The same file is in 18 other repositories
rmerk/crewai-headless-flow/.agents/skills/handoff Skill
1★ repoCompact the current conversation into a handoff document for another agent to pick up.
The same file is in 11 other repositories
looties-io/looties-skills/website-ai-agent-readiness Skill
1★ repoAudit or implement website readiness for AI agents and answer engines, including llms.txt, llms-full.txt, index.md, OKF/Open Knowledge Format bundles, machine-readable pricing files, ai-index.json, sitemap.xml, robots.txt, schema alignment, static asset headers, and SEO smoke tests. Use when the user mentions AI agent readiness, agent-readable site, llms.txt, OKF, Open Knowledge Format, knowledge bundle, AI-search readiness, answer engine readiness, machine-readable pricing, or agent-readable website assets.
mlava/scholar-sidekick-skills/skills/scholar-sidekick-mcp Skill
1★ repoUse the connected scholar-sidekick-mcp MCP server when the user mentions a scholarly identifier (DOI, PMID, PMCID, ISBN, arXiv, ISSN, NASA ADS bibcode, WHO IRIS URL) and wants structured metadata, a formatted citation, a bibliography export file, a retraction check, an open-access check, verification that a claimed citation is real (not fabricated), or a whole-bibliography audit (batch fabrication + retraction check). Requires the MCP server connected — it works anonymously, no API key required; for a zero-install path use the scholar-sidekick-api skill instead.
bcastelino/powerbi-dashboard-generator/skills/nlq-dashboard-orchestrator Skill
1★ repoTop-level entry point that turns a natural-language dashboard request into a complete Power BI Desktop Project (PBIP). This skill should be used whenever a user asks in plain English for a dashboard, report, or set of visuals without providing structured input. It introspects the data source, runs two explicit user-confirmation gates (data-model readiness and scaffold readiness), then orchestrates data-source-connector, query-to-pbip, theme-branding, and bi-dash-creator to produce the final dashboard.
wanghong5233/agent-engineering-kit/cursor/.cursor/skills/llm-cost-optimizer Skill
1★ repoOptimize LLM/Agent API cost without losing quality. Use when launching AI features, choosing models, prompt/context grows large, token spend rises, max_tokens/cache/model routing is unclear, or user asks to reduce LLM cost. Do NOT use for quality eval alone or RAG architecture alone.
wanghong5233/agent-engineering-kit/cursor/.cursor/skills/llm-observability-and-evals Skill
1★ repoDesign observability and eval contracts for production LLM/Agent systems. Use when adding ask/retrieval/tool/model flows, debugging quality regressions, defining trace/log/metric fields, building golden sets, or proving an Agent feature works. Do NOT use for cost-only work.
wanghong5233/agent-engineering-kit/cursor/.cursor/skills/tool-and-mcp-design Skill
1★ repoDesign production-grade Agent tools and MCP servers with clear schemas, permissions, errors, idempotency, and evals. Use when building tool calls, MCP integrations, external API actions, parser/web/search/database tools, or asking how an Agent should safely use external services.
wanghong5233/agent-engineering-kit/cursor/.cursor/skills/writing-engineering-playbook Skill
1★ repoWrite or refactor cross-project engineering playbook (Agent/LLM + backend; source in `docs/private/engineering-playbook/`, publish to 飞书/blog NOT GitHub). Use when user asks 写/沉淀/提炼 工程经验/Agent 经验/playbook/第一性原理/跨项目复用/技术博客/去AI味/AI痕迹. Do NOT use for project pitfalls.
wanghong5233/agent-engineering-kit/cursor/.cursor/skills/writing-skill Skill
1★ repoWrite or refactor an Agent Skill / SKILL.md (agentskills.io spec; portable across Cursor, Claude Code, etc.). Use when user asks to 写/create/refactor/audit/优化 a skill or SKILL.md, or asks why a skill 没触发/太长. Do NOT use for Cursor Rules, CLAUDE.md, AGENTS.md, MCP server.
frankxai/starlight-agent-skills/skills/substrate/agentic-income Skill
1★ repoThe substrate operating brain for building income systems with AI agents. Use when planning, building, or scaling an affiliate/content/product income network — deciding what to build next, where money actually comes from, how to make it compound, and how the system improves itself. Portable and brand-neutral. Composes affiliate-audit. Trigger phrases: build an income system, make money with agents, scale my content network, what should I build next, passive income with AI, monetize this site, agentic income.
frankxai/starlight-agent-skills/skills/substrate/swarm-queen-coordination Skill
1★ repoHow a stream queen coordinates a worker swarm and runs the escalation contract that keeps money safe — queens run streams, the founder owns capital, humans hold the final gate. Use when designing or operating a multi-agent income swarm, deciding what an agent may do autonomously vs. what it must escalate, or wiring the queen/worker/founder roles. Portable and brand-neutral. Trigger phrases: coordinate the swarm, stream queen, worker swarm, escalation contract, what can the agent decide, queen and workers, founder agent, agent autonomy limits.
anarefin/ai-agent-skills/prompt-master Skill
no license1★ repoGenerates optimized prompts for AI tools. Activates only when the user explicitly asks to write, fix, improve, or adapt a prompt for a specific AI tool (LLM, Cursor, Midjourney, image AI, video AI, coding agents, etc.). Does not activate for general conversation, coding tasks, document writing, or other non-prompt-engineering work.
The same file is in 3 other repositories
tdyzzsp47/claude-skills/skills/llm-integration Skill
1★ repoプロダクトへのLLM機能組み込みを体系的に進めるスキル。ユースケース適合性の判断からプロンプト設計・評価(eval)・コスト管理・セキュリティまでをカバーし、「動くが壊れやすい」実装を避ける。
tdyzzsp47/claude-skills/skills/orchestration Skill
1★ repo開発作業をモデル間で適切に分担するための司令塔プロトコル。メインモデル(司令塔)がタスクを分解し、Agent toolで適切なモデルのサブエージェントに委譲する。他のすべての開発系スキルの土台。複数ファイルの実装、大量の調査、定型作業を含むタスクで必ず参照する。
tdyzzsp47/claude-skills/skills/session-handoff Skill
1★ repo長時間タスクの作業状態をファイルに永続化し、セッション・担当者・モデルをまたいで30分で再開できる状態を保つスキル。中断・引き継ぎが発生するすべての場面で使う。
tdyzzsp47/claude-skills/skills/skill-authoring Skill
1★ repoClaude Code用スキル(SKILL.md)の新規作成・改善・品質検証を行うメタスキル。「新しいスキルを作りたい」「既存スキルを直したい」「スキルが発動しない」「このリポジトリの書き方規約を知りたい」と言われたときに使う。
nikhonit/llm-mentions-skills/skills/mentions-full Skill
1★ repoComplete AI brand-visibility toolkit via MentionsAPI.com. Four tools — check mentions/ranks/citations across ChatGPT, Claude, Gemini, Perplexity, Google AI Overviews/AI Mode, and Bing Copilot; discover queries to track; compare two brands or queries; and watch for changes over time.
k-anss/whetstone/skills/handoff-discipline Skill
1★ repoHand off the working state of one AI session to another that shares no context. Trigger when the user says "write a handoff for the next agent / another Claude / a fresh session", "brief the next instance", "sync state to my other project", or "carry this over without losing what's decided". Produces a self-contained handoff doc that separates what must NOT be reopened from current status from strategic rationale, and defines the receiving agent's role. Do NOT trigger for: summarizing a document for a human reader, writing meeting notes, or compressing a conversation for the same ongoing session (that is summarization, not handoff).
Rookage/agent-discipline/skills/session-handoff Skill
1★ repoUse when preparing a handoff for another session or agent, summarizing current progress, preserving context, avoiding repeated questions, or ending a long task. Creates a concise handoff document with goals, state, completed work, decisions, next steps, and risks.
roronoazoroshao369/vibe-coding-os/skills/agents/architect-agent Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/agents/implementer-agent Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/agents/reviewer-agent Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/agents/tester-agent Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/core/adaptive-prompt-selection Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/core/claude-code-hooks-pack Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/core/model-weakness-memory Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/core/orchestration-workflows Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/core/prompt-architecture Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/core/subagent-driven-development Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/core/superagent-orchestration Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/core/team-agent-orchestration Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/memory/memory-compression Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/memory/memory-ingestion Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/memory/memory-search Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/memory/project-memory Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
roronoazoroshao369/vibe-coding-os/skills/templates/agents-md-compact Skill
1★ repoVibe Coding OS — Claude/Codex/Cursor skill framework with 139 skills, 111 commands, 95 templates, 22 tracked sources, 28/28 validation gates PASS. Quality Shield, Engineering Discipline Pack, plugin marketplace.
jlreyes/swift-agent-kit/skills/device-interaction Skill
no license1★ repoApple's protocol for verifying iOS app behavior on a simulator or device: session lifecycle, capturing screenshots and UI hierarchies, synthesizing taps, swipes, and typing via hierarchy center coordinates, and judging functional vs visual vs transient issues. Use when verifying that UI changes work on a device, debugging touch issues, or asked 'does it work'. Requires the xcode-tools MCP server and an iOS 27+ simulator runtime. (Content extracted from your local Xcode on install.)
jlreyes/swift-agent-kit/skills/translation-coordinator Skill
no license1★ repoOrchestrates Xcode String Catalog localization: prepares the project via LocalizationPlanner, fetches untranslated keys by state, and delegates batches to translator subagents that use the translation skill. Use when asked to localize an Xcode project or translate String Catalogs. Requires the xcode-tools MCP server. (Content extracted from your local Xcode on install.)
jlreyes/swift-agent-kit/skills/translation Skill
no license1★ repoTranslates a provided list of Xcode String Catalog keys into one provided target locale via the StringCatalog MCP tools, with per-key context lookup. Worker role inside a localization flow — only use when delegated an explicit key list and target locale (normally by translation-coordinator); it aborts without them. Requires the xcode-tools MCP server. (Content extracted from your local Xcode on install.)
jlreyes/swift-agent-kit/skills/xcode-tools Skill
no license1★ repoOperating guide for Xcode 27's bundled MCP server (mcpbridge): when to use each mcp__xcode-tools__* tool versus the agent's native tools or the CLI, connection setup and repair, output-noise control, and the tools that are genuinely unique (live per-file diagnostics, the shared lldb session, preview rendering, project-membership file operations, device driving). Use whenever xcode-tools MCP tools are available and an Xcode project is involved — including building, running, testing, debugging, rendering previews, or driving the simulator — before reaching for any mcp__xcode-tools__* tool, or when those tools are missing or erroring.
balacodeio/balakit/skills/everything-seo Skill
1★ repoExpert SEO advisor for modern search optimization. Covers technical SEO audits, crawlability, semantic SEO, E-E-A-T, digital PR, GEO (Generative Engine Optimization), RAG optimization, and multi-engine (Google + Bing) strategy. Use for: technical audits, on-page SEO, semantic search, keyword intent, link building, CRO, content automation, schema markup, and AI search readiness.
balacodeio/balakit/skills/nlm-skill Skill
1★ repoExpert guide for the NotebookLM CLI (nlm) and MCP server. Covers authentication, notebook management, source addition (URLs, YouTube, text, Drive, files), content generation (audio, video, report, quiz, flashcards, mind map, slides, infographic, data table), research, artifact management, and automation workflows. Progressive disclosure: body carries critical rules + decision tree + phase workflow; full command reference in references/. Triggers on "nlm", "notebooklm", "notebook lm", "podcast generation", "audio overview", or any NotebookLM automation task.
balacodeio/balakit/.cursor/skills/authoring-skills-and-rules Skill
1★ repoAuthor Skills (SKILL.md) and rules/instructions across Claude Code, Cursor, OpenCode, Codex, and GitHub Copilot. Covers frontmatter, file layout, activation model, cross-platform mirroring, and the universal craft: progressive disclosure, trigger-rich descriptions (context pointers), leading words, token budget, naming, and what to leave out. Use when the user asks to write, scaffold, refactor, port, or review a Skill, Cursor rule (.mdc), AGENTS.md, copilot-instructions, or prompt file — or to make one Skill/rule work across multiple agents.
balacodeio/balakit/.cursor/skills/everything-seo Skill
1★ repoExpert SEO advisor for modern search optimization. Covers technical SEO audits, crawlability, semantic SEO, E-E-A-T, digital PR, GEO (Generative Engine Optimization), RAG optimization, and multi-engine (Google + Bing) strategy. Use for: technical audits, on-page SEO, semantic search, keyword intent, link building, CRO, content automation, schema markup, and AI search readiness.
balacodeio/balakit/.cursor/skills/nlm-skill Skill
1★ repoExpert guide for the NotebookLM CLI (nlm) and MCP server. Covers authentication, notebook management, source addition (URLs, YouTube, text, Drive, files), content generation (audio, video, report, quiz, flashcards, mind map, slides, infographic, data table), research, artifact management, and automation workflows. Progressive disclosure: body carries critical rules + decision tree + phase workflow; full command reference in references/. Triggers on "nlm", "notebooklm", "notebook lm", "podcast generation", "audio overview", or any NotebookLM automation task.
quantum-box/agent-packages/plugins/tachyon/skills/tachyon-browser Skill
no license1★ repoUse the Tachyon k3s Cloudflare Mesh remote Chromium CDP gateway with agent-browser, Playwright, Codex, or Claude Code. Trigger when a task needs a browser reachable from Sakura VPS or another Mesh client, remote CDP automation, temporary browser sessions, named browser sessions, or AgentCore Browser-like session-per-Pod browser control through browser-cdp.intra.quantum-box.com.
MarioJames/skill-foundry/skills/workspace-knowledge-graph Skill
1★ repoProduction-proven agent skills: agent-swarm, asset-validation, browser-harness, awesome-presentation, workspace-knowledge-graph.
Pipeline orchestration and data quality
scumunna/programmatic-skills/skills/pipeline-orchestration-and-data-quality Skill
1★ repoBuild idempotent extraction, load, and transform pipelines that pull marketing data (GA4, CM360, DV360, DSPs, retail media) into a warehouse on a schedule, with watermarks, freshness SLAs, schema-drift detection, and tests as code so a rerun never double counts and a broken feed never reaches a dashboard silently. Use when the user asks how to schedule and orchestrate a marketing data pipeline, make a load idempotent, backfill without duplicates, set incremental watermarks, detect schema drift, add data-quality tests, set a freshness SLA, handle late-arriving data, paginate an ad API with backoff, or why last night's ingestion silently produced stale or partial numbers.