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
MustaphaSteph/agent-bus-plugins/skills/agent-bus-listener Skill
1★ repoUse when this session should become an agent-bus helper/listener that quietly waits for team-scoped messages or tasks, answers when called, and keeps its task/status current.
MustaphaSteph/agent-bus-plugins/skills/agent-bus-read-first Skill
1★ repoSTART HERE when a user wants multiple AI coding sessions to coordinate through agent-bus, install the plugin, create a team, use the web cockpit, or decide whether to use MCP tools, CLI commands, listener sessions, tasks, memories, reviews, or final reports.
MustaphaSteph/agent-bus-plugins/skills/agent-bus-workflows Skill
1★ repoUse when planning or enforcing agent-bus software-team workflows: backlog, Kanban, memory loop, verifier gate, task ownership, review approvals, and final merge reports.
MustaphaSteph/agent-bus-plugins/skills/agent-bus Skill
1★ repoCoordinate work across Claude/Codex/Cursor sessions on the same machine via a local message bus. Use to delegate to helpers, get a second opinion, ask specialists by capability, or track shared tasks.
krkn-s/ai-workspace/skills/optimize-agent-instructions Skill
1★ repoLean audit and rewrite workflow for agent instruction files such as AGENTS.md, CLAUDE.md, system prompts, rules files, and skill instructions. Use when the agent needs to clean up, shorten, or refactor an instruction file, remove redundant or contradictory rules, reduce unnecessary planning, questions, or validation steps, fix a CLAUDE.md or AGENTS.md that feels too long or too strict, review a skill's instructions for leanness, or consolidate overlapping project rules into a minimal high-signal replacement. Déclenche aussi sur les requêtes en français, par exemple alléger, raccourcir, nettoyer ou réécrire un fichier d'instructions (AGENTS.md, CLAUDE.md, prompt système, fichier de règles, skill), supprimer les règles redondantes ou contradictoires, simplifier un CLAUDE.md ou AGENTS.md trop long ou trop strict, ou consolider des règles projet qui se chevauchent en un fichier minimal.
niftymonkey/skills/skills/externalize Skill
1★ repoUse as a standing disposition to write hard-won, expensive, non-reproducible context to the work-thread handoff file the instant it is earned during work, so it can never die in the context window or get lossily summarized by compaction. The continuous-write capture sibling of work-router. Fires when knowledge is earned, not at session end and not at a context-window threshold.
agentjido/jido-skills/.agents/skills/jido-agent Skill
1★ repoBuild Jido Agents with schema validation, cmd/2 command pattern, plugins, and directives. Use when creating new agents, defining agent schemas, adding plugins, or working with the AgentServer runtime.
agentjido/jido-skills/.cursor/skills/jido-agent Skill
1★ repoBuild Jido Agents with schema validation, cmd/2 command pattern, plugins, and directives. Use when creating new agents, defining agent schemas, adding plugins, or working with the AgentServer runtime.
agentjido/jido-skills/.gemini/skills/jido-agent Skill
1★ repoBuild Jido Agents with schema validation, cmd/2 command pattern, plugins, and directives. Use when creating new agents, defining agent schemas, adding plugins, or working with the AgentServer runtime.
agentjido/jido-skills/source/skills/jido-agent Skill
1★ repoBuild and maintain Jido agent modules, schemas, command handling, plugins, directives, and runtime wiring. Use when creating agents, changing agent behavior, or working with AgentServer-style runtime integration.
pjankiewicz/skills/skills/open-multi-agent Skill
1★ repoUse when building autonomous agent systems in TypeScript/Node.js — triaging a list (GitHub issues, files, URLs) and dispatching parallel workers per item, running each worker inside its own git worktree or sandbox, fanning out one ephemeral agent per dynamic input via AgentPool, decomposing a goal into a parallel task DAG, running local gemma/qwen via Ollama or vLLM with concurrency caps, mixing providers (Anthropic/OpenAI/Gemini/Grok/DeepSeek/local) in one team, connecting MCP servers, or producing Zod-validated structured output. Triggers on `@jackchen_me/open-multi-agent`, `OpenMultiAgent`, `AgentPool`, `runTeam`, `runTasks`, `runParallel`, `runEphemeral`, the `oma` CLI, "agent farm", "parallel workers per issue", "TypeScript multi-agent framework". Skip for single-shot LLM calls, Python multi-agent stacks, or LangGraph/Mastra/CrewAI work.
GerardoRdz96/aios-starter-kit/.claude/skills/hooks-builder Skill
1★ repoUse when the user asks to design, build, optimize, or audit Claude Code HOOKS — event-driven shell commands in settings.json that fire on SessionStart, PreToolUse, PostToolUse, Stop, etc. Triggers — "build a hook", "add a hook", "run X every time I edit/commit/start a session", "block tool X when…", "audit my hooks", or `/hooks-builder`. Sibling of `/skill-builder`, `/agent-builder`, `/routines-builder`, `/agents-team-builder`, `/plugin-builder` — this one is the event-driven-local specialist (the hooks lane `/routines-builder` hands off to).
GerardoRdz96/aios-starter-kit/.claude/skills/plugin-builder Skill
1★ repoUse when the user asks to build, package, bundle, optimize, or audit a Claude Code plugin or marketplace — a distributable bundle of skills + agents + commands + hooks installed via `/plugin`. Triggers — "build a plugin", "package this as a plugin", "bundle my skills", "make a marketplace", "turn these skills into something shareable", or `/plugin-builder`. The packaging + distribution sibling of `/skill-builder`, `/agent-builder`, `/agents-team-builder`, `/routines-builder`, `/hooks-builder`, `/workflow-builder`. Runs a decision gate and Discovery Interview before writing files.
GerardoRdz96/aios-starter-kit/.claude/skills/routines-builder Skill
1★ repoUse when someone wants to set up recurring/automated cadence — build, optimize, or audit a Claude Code Routine (cloud scheduled agent via /schedule), a local hook, a /loop, or a daily-*/weekly-* ritual skill. Triggers on "schedule this", "automate X weekly", "make my AIOS run itself", "wire a recurring job", "attack our cadence gap". Runs a decision gate and Discovery Interview, then a mandatory supervised test before arming anything.
GerardoRdz96/aios-starter-kit/.claude/skills/session-handoff Skill
1★ repoUse when the user wants to wrap up, clear context, switch machines or tools, or continue work in a fresh session — "hand off", "wrap up this session", "save where we are", "I need to clear context", "write a handoff", "/session-handoff". Prints and saves a structured handoff note (what we did, files touched, open decisions, next steps) so the next session — in any tool — can pick up exactly where this one left off.
GerardoRdz96/aios-starter-kit/.claude/skills/skill-builder Skill
1★ repoUse when someone asks to build, create, make, optimize, or audit a Claude Code skill -- "build me a skill", "make a /slash-command", "create a new skill", "audit my skills", or /skill-builder. Runs a Discovery Interview before writing files.
GerardoRdz96/aios-starter-kit/.claude/skills/workflow-builder Skill
1★ repoUse when the user asks to design, build, optimize, or audit a SAVED DYNAMIC WORKFLOW — a JS orchestration file in `.claude/workflows/` that fans work out to N parallel sub-agents and merges results. Triggers — "build a workflow", "save this as a workflow", "make this rerunnable as a workflow", "turn this fan-out into a saved workflow", "audit my workflows", or `/workflow-builder`. Sibling of `/skill-builder`, `/agent-builder`, `/routines-builder`, `/agents-team-builder`, `/plugin-builder`, `/hooks-builder` — this one is the width-orchestration specialist. Builds only; it never fires the workflow for you.
fbarrento/laravel-agent-kit/.agents/skills/handoff Skill
no license1★ repoCompact the current conversation into a handoff document for another agent to pick up.
The same file is in 13 other repositories
fbarrento/laravel-agent-kit/.agents/skills/setup-matt-pocock-skills Skill
no license1★ repoSets up an `## Agent skills` block in AGENTS.md/CLAUDE.md and `docs/agents/` so the engineering skills know this repo's issue tracker (GitHub or local markdown), triage label vocabulary, and domain doc layout. Run before first use of `to-issues`, `to-prd`, `triage`, `diagnose`, `tdd`, `improve-codebase-architecture`, or `zoom-out` — or if those skills appear to be missing context about the issue tracker, triage labels, or domain docs.
The same file is in 15 other repositories
fbarrento/laravel-agent-kit/.agents/skills/write-a-skill Skill
no license1★ repoCreate new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill.
The same file is in 20 other repositories
nianbaizy/grad-agent-kit/templates/skill-template Skill
1★ repoReusable AI agent skills for research writing, experiment analysis, academic presentation, reviewer simulation, and project delivery.
andrey-learning-machines/swe-harness/plugins/swe-harness/skills/mcp-server-builder Skill
1★ repoUse when designing, implementing, deploying, or reviewing Model Context Protocol (MCP) servers, especially remote HTTP MCP servers, tool schemas, write-capable tools, connector metadata, authentication, authorization, prompt-injection defenses, audit logging, deployment smoke tests, or quality assurance for MCP integrations.
andrey-learning-machines/swe-harness/plugins/swe-harness/skills/unicorn-orchestrator Skill
1★ repoCodex-native orchestration workflow for the Unicorn Team custom agents. Trigger when the user explicitly asks for delegation, subagents, or parallel agent work and the task benefits from specialized architecture, implementation, review, DevOps, language-ramp-up, or UI/UX help. Use for multi-step coding work, architecture planning, review pipelines, deployment prep, or mixed UI/backend efforts where bounded child-agent tasks materially help.
scwlkr/skillsies/skills/write-a-skill Skill
no license1★ repoCreate new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill.
The same file is in 1 other repository
VdustR/skills/skills/vp-foreign-agent-skill-loader Skill
1★ repoReusable Agent Skills by VdustR
pavelsimo/skills/skills/create-skill Skill
no license1★ repoBootstraps a new agent skill by generating SKILL.md, README.md, LICENSE, and optional reference files as a directory inside the pavelsimo/skills repo and registering it in the index README. Use when the user wants to create a new Claude Code skill from scratch.
Soham407/studio-kickstart/architecture/claude-plugin-powerpack Skill
1★ repoInstalls and documents the recommended Claude Code plugin stack for a high-leverage open-source agent setup. Use when setting up Claude plugins, when the user mentions skill-creator, superpowers, context-mode, claude-mem, frontend-design, get-shit-done-cc, or asks for the Studio Claude plugin pack.
Soham407/studio-kickstart/architecture/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
Soham407/studio-kickstart/architecture/setup-matt-pocock-skills Skill
1★ repoSets up an `## Agent skills` block in AGENTS.md/CLAUDE.md and `docs/agents/` so the engineering skills know this repo's issue tracker (GitHub or local markdown), triage label vocabulary, and domain doc layout. Run before first use of `to-issues`, `to-prd`, `triage`, `diagnose`, `tdd`, `improve-codebase-architecture`, or `zoom-out` — or if those skills appear to be missing context about the issue tracker, triage labels, or domain docs.
The same file is in 15 other repositories
Soham407/studio-kickstart/architecture/write-a-skill Skill
1★ repoCreate new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill.
The same file is in 20 other repositories
Soham407/studio-kickstart/business/agentic-seo Skill
1★ repoUse when auditing or improving documentation and websites for discoverability, parseability, and usefulness to AI coding agents.
Mobile maestro e2e orchestrator
mo-hawary/hawary-workflow-skills/skills/mobile-maestro-e2e-orchestrator Skill
1★ repoRun evidence-driven mobile E2E tests with Maestro, device logs, checkpoints, and targeted retries for iOS or Android flows.
HKTITAN/cursor-sdk/skills/cursor-mcp Skill
1★ repoConfigure MCP servers on Cursor SDK agents — stdio subprocess, HTTP, and SSE transports, inline vs file-based config, and per-run overrides. Use when extending an agent's tools beyond the built-ins.
Ts tsconfig modern strict starter
kongyo2/agent-primary-ts-starters-src/skills/ts-tsconfig-modern-strict-starter Skill
1★ repoSet up a TypeScript tsconfig.json with modern (ESNext + bundler/nodenext) and maximally strict settings, tuned so that `tsc --noEmit` becomes a fast, reliable inner feedback loop for the LLM agent doing the editing. Use whenever a tsconfig.json is being created, edited, reviewed, troubleshot, or even just discussed — apply the standing policy even when the user only asks about a single option. Default to this skill instead of free-recall whenever the answer involves a tsconfig field.
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/build-agent-harnesses Skill
1★ repoRuns checklists and workflows for designing, hardening, and operating agent harnesses — guardrails, verify steps, tool loops, durable sessions, eval, quotas, coding/deep-research/voice harnesses. Use when the user builds agent runtimes, asks what a harness is, scopes agent vs workflow, or says "agent harness", "guardrails", "verify step", "FOMAT", "durable session".
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/build-aiewf-2024-rag Skill
1★ repoRuns workflows for AIEWF 2024 RAG talks — layout-aware ingest, GraphRAG, eval-driven tuning, Pydantic structured IO, MongoDB vectors, EMT memory, construction vertical agents, personal-assistant data graphs. Use when the user says "RAG", "GraphRAG", "chunking PDF", "Pydantic RAG", or AIEWF retrieval playlist.
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/build-rag-search-stacks Skill
1★ repoRuns checklists and workflows for designing, layering, evaluating, and operating production RAG and AI search — hybrid retrieval, agentic context tools, enterprise eval, web neural search, managed platforms, document agents. Use when the user builds retrieval for agents, picks vector vs lexical vs hybrid, designs eval for augmented AI, scales regulated RAG, compares search APIs, or says "RAG stack", "hybrid search", "agentic RAG", "vector benchmark".
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/build-with-microsoft-ai Skill
1★ repoRuns workflows for Microsoft agent stacks — agent observability gaps, eval-driven alignment, and local vs background vs cloud agents in VS Code/Copilot. Use when the user builds on Azure Foundry, GitHub Copilot agents, agent evals, or says "mind the gap observability", "Copilot worktrees", "background agent".
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/build-with-openai Skill
1★ repoRuns OpenAI API workflows for agents, function calling, voice/Realtime, fine-tuning (SFT/DPO/RFT), sandboxes, Codex coding agents, and multimodal products. Use when building with OpenAI Agents SDK, Realtime API, tool loops, enterprise agent rollout, microVM sandboxes, or GPT-4o vision/voice.
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/design-agent-skills Skill
1★ repoRuns workflows for authoring agent skills — SKILL.md structure, MCP pairing, Langfuse lessons, progressive disclosure, validation. Use when the user writes skills, combines MCP+skills, or says "agent skills", "SKILL.md", "context gap".
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/evaluate-with-braintrust Skill
1★ repoEvaluates LLM and agent products with Braintrust—offline/online evals, SDK/CI workflows, eval platform design, Zapier/Notion product ops, and Loop-style optimization. Use when the user mentions Braintrust, eval playground, autoevals, production log evals, or agent quality platforms.
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/lead-ai-engineering-teams Skill
1★ repoRuns leadership workflows for AI-era engineering — context CDLC, org-wide AI adoption, agent-first throughput, and post-engineer team design. Use when the user leads eng org change, measures AI productivity, or says "double throughput", "context is the new code", "agents don't do standups", "AI SDLC rollout".
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/learn-summit-2023-remote Skill
1★ repoApplies AI Engineer Summit 2023 remote-talk playbooks for evals, hybrid RAG grounding, model selection, fine-tuning, prompt ops, multimodal apps, coding-agent maturity, and career transition from fullstack. Use when reproducing summit remote lessons, comparing LLM eval tools, or scoping hybrid retrieval; or when the user says "summit 2023 remote", "llmeval", "hybrid context query", "AI maturity model".
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/observe-ai-production Skill
1★ repoObserves and improves production LLM apps with Arize Phoenix—tracing, layered agent evals, prompt learning loops, and PM/engineering eval pipelines. Use when the user mentions Arize, Phoenix, OpenInference, agent router evals, prompt optimization from traces, or shipping AI with production observability.
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/operate-agent-memory Skill
1★ repoRuns checklists and workflows for agent context engineering — separating context from memory, smart truncation, sub-agent offload, and long-session evals. Use when the user hits context limits on observability or trace-heavy agents, designs multi-turn memory, or says "context management", "escape context window", "agent forgets follow-ups".
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/operate-openclaw-agents Skill
1★ repoRuns checklists and workflows for deploying, securing, scaling, and evaluating OpenClaw agents — personal incremental setup, enterprise containers, Kubernetes, maintainer PR triage, malleable evals, local vs cloud, team Slack agents. Use when the user deploys OpenClaw, hardens a rollout, designs agent evals, sets up a personal agent, compares Viktor-style Slack agents vs OpenClaw, or says "OpenClaw K8s", "OpenClaw security", "agent eval harness".
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/optimize-llm-inference Skill
1★ repoRuns workflows for GPU/inference talks — Groq LPU latency, OSS inference stacks, MoE networking, serverless fine-tune, frontier cluster design, Devin agent infra, Gemma sizing. Use when optimizing inference $/token, training networks, or picking open models.
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/run-aiewf-2024-evals Skill
1★ repoRuns checklists for AIEWF 2024 eval and LLM ops talks — domain eval ladders, LLM judges, enterprise deploy, Zapier+Braintrust loops, GenAI maturity before fine-tune. Use when the user says "evals", "LLM judge", "fine-tune decision", "enterprise GenAI deploy", or AIEWF eval playlist.
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/run-llm-evals Skill
1★ repoRuns checklists and workflows for designing, operating, and improving LLM evals — domain harnesses, agent/voice evals, LLM judges, metrics workshops, product flywheels, dynamic search eval, enterprise conversation intelligence. Use when the user says "evals", "LLM judge", "eval harness", "regression on prompts", "agent eval", "eval dataset", or needs to ship with measurement before prod.
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/run-summit-2023-workshops Skill
1★ repoRuns AI Engineer Summit 2023 workshop playbooks — foundations (101), inference (201), product patterns beyond inference (Rest of the Owl), and Anthropic prompt engineering principles. Use when onboarding to AI engineering, designing inference stacks, or improving task prompts; or when the user says "summit 2023 workshop", "AI engineering 101", "inference workshop", "prompt principles Karina".
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/ship-aiewf-2024-agents Skill
1★ repoRuns workflows for AIEWF 2024 agent talks — Vertex governance, Minecraft JSON actions, voice pipelines, agentic commerce, personas, CrewAI dogfooding, job automation realism. Use when shipping agents, voice agents, or agent marketplaces.
zrosenbauer/skills/skills/skill-creator Skill
1★ repoThis skill should be used when the user wants to create a new agent skill, scaffold a SKILL.md, validate an existing skill against repo rules, or refactor a skill to match this monorepo's conventions. Common triggers include "build a skill for X", "create a new skill", "scaffold a skill", "add a skill that does Y", "make me a skill", "audit this skill against our rules", and "refactor this skill to match repo conventions". Enforces kebab-case naming, verbatim trigger phrases, and selective XML for example boundaries. Skip when modifying source code, debugging an existing skill, or writing non-skill markdown.
mattdweigand-sketch/agent-skills/skills/context-window-audit Skill
1★ repoAudit an agent setup for token waste, context-window bloat, stale instructions, and dead command wrappers. Use when the user says context-window audit, context audit, audit my context, check my settings, why is Codex so slow, token optimization, or asks why startup context is too large.
mattdweigand-sketch/agent-skills/skills/n-agentic-harnesses Skill
1★ repoDesign, evaluate, and improve agentic harnesses — the orchestration layer around LLM-powered tools, agents, assistants, copilots, workflow runtimes, and AI-driven product features. Use this skill whenever the user mentions building an agentic system, structuring tool use, adding permissions or approval gates, designing multi-step AI workflows, managing context windows or memory, making agents durable or resumable, evaluating or pressure-testing an existing harness, planning phased implementation for an AI product, reviewing agent architecture, improving agent UX or observability, or asking how to know if their harness is actually good. Also use when the user describes problems that imply harness gaps — like agents doing unexpected things, context getting stale, sessions not surviving crashes, tools running without permission, or costs spiraling — even if they do not use the word "harness."
mattdweigand-sketch/agent-skills/skills/quality-check-eval Skill
1★ repoSort AI output quality checks into four buckets: automate in code, judge with an LLM, keep for human review, or remove. Use when reviews are too manual, evals are vague, a workflow needs clearer quality gates, or the user wants to decide which AI output checks are worth automating. Not for checking whether a system grades encoded judgment against real outcomes; use cyborg-check for that.
mattdweigand-sketch/agent-skills/skills/ralf-loop Skill
1★ repoRun a repo-agnostic RALF-style improvement loop: run a smoke test or workflow, analyze failures against frozen gates, make one bounded fix, review the change, rerun, and stop at a clear terminal point. Use when the user says "RALF loop", "loop testing", "multi-agent test loop", "submit/review/analyze/redeploy", "tighten this system with agents", "run the smoke test until it converges", or asks to repeatedly harden a repo, harness, skill, prompt, test suite, workflow, retrieval system, exporter, parser, or agentic product using local tests and review gates. Works on any repo with a local command or scorable workflow.
mattdweigand-sketch/agent-skills/skills/skill-creator-v3 Skill
1★ repoCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
mattdweigand-sketch/agent-skills/skills/skill-tune Skill
1★ repoAudit or refactor a skill or prompt artifact for prompt technical debt. Use when explicitly asked to review SKILL.md, AGENTS.md, CLAUDE.md, system prompts, command wrappers, or tool instructions for prompt debt, over-steering, context bloat, stale model-specific instructions, duplicated canonical rules, misplaced executable checks, or missing authority boundaries. Use as a debt review pass before or after skill creation; use skill-creator for net-new skill scaffolding.
mattdweigand-sketch/agent-skills/skills/swarm Skill
1★ repoInvocation-only workflow. Use only when the user includes literal `swarm` as an instruction to apply this workflow. Do not trigger for broader agent, harness, judge, checker, orchestration, or multi-agent requests unless the literal `swarm` invocation is present.
viditkbhatnagar/immunize/src/immunize/patterns/react-hook-missing-dep Skill
1★ repoUse when writing React components with useEffect or useCallback to ensure every reactive value referenced in the hook body appears in the dependency array.