Context & AI engineering
25,841 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
Harduex/agentic-notebook/skills/agentic-notebook Skill
1★ repoTurn the agent into a NotebookLM-style, source-grounded research notebook over a local folder of files (PDF books, papers, notes, docx, pptx, epub, HTML, CSV, transcripts, images). Use this skill whenever the user says "act like NotebookLM / Gemini Notebook", "chat with my documents", "answer only from these files", asks questions about a folder of PDFs or books, or requests NotebookLM-style artifacts from their own sources: notebook guide, briefing doc, study guide, FAQ, timeline, mind map, flashcards, quiz, data table, audio overview / podcast script, video overview, slide deck, or infographic. Also use it for "summarize everything in this folder", literature reviews over local papers, building a study workflow from course materials, or any task where answers must stay strictly grounded in user-provided documents with verifiable citations — even if the user never says the word "NotebookLM".
baldaworks/hostplugin/plugins/hostplugin/prefixed-skills/hostplugin-author Skill
1★ repoAuthor, audit, or update portable plugins for Codex, Claude Code, Grok Build, GitHub Copilot CLI, OpenCode, and Cursor. Use when an agent must inspect a project or CLI, design host-native skills, commands, agents, hooks, MCP/LSP integrations, rules, or apps, scaffold either a standalone marketplace or an embedded plugin, and preview every file before writing.
Hyper-AI-Lab/agentic-design/skills/agentic-design Skill
1★ repoDesigns and implements distinctive, production-ready web interfaces grounded in product context, brand, content, and user goals. Use when creating or redesigning websites, landing pages, web apps, dashboards, components, or frontend visual systems, especially when the result should avoid generic AI-generated patterns.
Zhishuo build pdf knowledge base
zgzdhr/zhishuo-workflows/skills/zhishuo-build-pdf-knowledge-base Skill
1★ repoUse when the user needs to build a durable searchable knowledge base from multiple PDFs, a book-length PDF, manuals, proceedings, or mixed scanned references. Use MinerU as the required parser for this formal knowledge-base branch. Do not use for one-off reading or extracting a small number of ordinary PDFs.
fullREFIT/skills/skills/1pw-env Skill
1★ repoManage credentials across all projects using the 1Password environment variable system. Handles op run zero-disk secret injection, syncing credentials from 1Password vault to shell environment, adding new credentials, service account setup for AI agents, SSH agent integration, SDK usage, rate limit awareness, and troubleshooting auth failures. Covers: 1Password vault → op run → subprocess (preferred) and vault → op CLI → ~/.env.1p.tpl → ~/.env.mcp → shell (fallback). MANDATORY TRIGGERS: 1password, credentials, env vars, environment variables, API key missing, auth failure, credential sync, op CLI, secrets management, .env.mcp, credential setup, new credential, Touch ID timeout, vault, shell plugins, biometric auth, service account, SDK, op run, SSH agent, rate limit, Connect Server, Environments, agentic autofill, OP_SERVICE_ACCOUNT_TOKEN, op run wrapper.
authenticfake/clike/extensions/vscode/templates/harper-init/.clike/skills/ai-rag-eval-guardrails Skill
1★ repoGuardrails for AI/RAG features: provenance, prompt-injection defense, output validation before business action, and eval evidence.
MatrixFounder/Universal-skills/archive/skill-creator-ref Skill
1★ repoGuide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
The same file is in 19 other repositories
partoska/p6a-skills/skills/partoska Skill
1★ repoHelp users manage Partoska.com photo-sharing events through the Partoska MCP server — find and create events, browse/preview/upload/curate photos, approve moderated uploads, share via invite links, QR codes, and printable cards, and design event assets. Use when Partoska MCP tools (e.g. event-query, photo-list) are available in the session.
Eng-Muhammad-Affan/digital-fte/.qwen/skills/email-sending Skill
no license1★ repoSend emails via Gmail using the Email MCP server. Supports sending, drafting, searching emails, and marking as read. Use when tasks require email communication, sending invoices, replying to clients, or managing Gmail inbox.
bingooyong/agent-skills/skills/agents-md-maintainer Skill
no license1★ repoMaintain AGENTS.md as the stable, repo-level rule file for AI agents. Use when (1) adding or updating project constraints, build commands, test commands, or collaboration rules, (2) auditing AGENTS.md for stale or misplaced content, (3) ensuring only durable rules live in AGENTS.md while ephemeral context stays elsewhere. Prevents polluting AGENTS.md with temporary session context.
Aliciawque/openclaw-personalized-agent-forge/02-professional-agent-forge Skill
no license1★ repoBuild a complete OpenClaw agent package for a real profession or job role. Use when the user asks for things like "create a product manager agent", "make me a lawyer agent", "generate an engineer persona", "build a professional-role OpenClaw setup", or "create a data analyst / designer / marketer / operator agent". Produce a role-specific package centered on `soul.md`, `identity.md`, `memory.md`, `agents.md`, and `tools.md`, plus a recommended supporting-skill stack.
corticalstack/flow/.claude/skills/create-handoff Skill
1★ repoWrite a structured handoff document under flow/handoffs/ to transfer work to another session. Explicit-only; run only when invoked via /create-handoff or the user asks for a handoff.
jtmthf/skills/skills/agents-md Skill
no license1★ repoAuthor or improve AGENTS.md / CLAUDE.md agent-instruction files — a lean root, progressive disclosure via nested files, conditional gating, the vendor-neutral AGENTS.md + CLAUDE.md setup, and grounding the file in real pain points mined from Claude Code session logs.
coreline-ai/android-subagent-skill/skills/document-reviewer-agent Skill
1★ repo[Android App Development] [Claude Code] 안드로이드 최초 기획 문서(PRD, TRD 등)를 리뷰하고 안드로이드 생태계 제약사항에 맞춰 정합성을 교정, 문서를 상세화하는 안드로이드 메인 Agent 스킬입니다.
gabeosx/agentskillsdir/.gemini/skills/skill-creator Skill
no license1★ repoGuide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
corneliu-iancu/skills/skills/meta/subagent-driven-development Skill
no license1★ repoUse when executing implementation plans with independent tasks in the current session
SepantaJoshani/agents-skills/skills/codex-open-worktree Skill
1★ repoOpen any existing Git-registered worktree as a workspace in the macOS Codex desktop app, including installations branded as ChatGPT. Use when the user asks to open, switch to, move to, or continue in an existing worktree from Codex; invokes `$codex-open-worktree`; supplies a branch name, worktree directory, or worktree path; or wants Codex to discover a repository's worktrees and ask which one to open. Works with worktrees created by Git, Claude, Codex, Supacode, or other tools at any path. Opens a workspace without creating, removing, checking out, or modifying a worktree or branch.
plaited/development-skills/skills/optimize-agents-md Skill
1★ repoOptimize AGENTS.md and rules for token efficiency. Auto-invoked when user asks about improving agent instructions, compressing AGENTS.md, or making rules more effective.
NecturaLabs/AgentSkills/skills/agents-md-manager Skill
1★ repoCreate or update the project's AGENTS.md file based on codebase analysis. Invoke manually via /agents-md-manager, or triggered automatically as the final step of superpowers plan execution.
trsdn/github-copilot-cli/.github/skills/copilot-setup-audit Skill
1★ repoAudit a repository's Copilot CLI customization setup and suggest improvements. Use when the user wants to review their Copilot configuration, find gaps, validate agents, skills, instructions, hooks, plugins, or MCP setup, or optimize their CLI customization setup.
ulpi-io/plugin-marketplace/plugins/404kidwiz/skills/agent-organizer Skill
no license1★ repoExpert in designing, orchestrating, and managing multi-agent systems (MAS). Specializes in agent collaboration patterns, hierarchical structures, and swarm intelligence. Use when building agent teams, designing agent communication, or orchestrating autonomous workflows.
eikonoikari1/skillport/.cursor/skills/skillport Skill
1★ repoUniversal skill/rule adapter across AI coding harnesses. Converts skills, rules, and instructions between Claude Code, Cursor, Codex CLI, OpenClaw, Copilot, and Windsurf. Handles harness-specific features (hooks, subagents, allowed-tools, globs, dynamic context) with functional equivalents or annotated fallbacks. Use when asked to "port", "convert", "adapt", "translate", or "make work in" a different harness. Also use when asked to "detect" harness configs in a project.
zinan92/repo-evals/repos/anthropics--skill-creator/runs/2026-05-05/run-static-checks/artifacts 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.
The same file is in 66 other repositories
NinetrixAI/skills-hub/skills/agent-browser Skill
no license1★ repoTeach agents to automate web browsers via agent-browser CLI — navigate, snapshot refs, interact, extract data, handle auth
rnd-pro/stitch-skills/skills/enhance-prompt Skill
1★ repoTransforms vague UI ideas into polished, Stitch-optimized prompts with CSS custom properties and semantic design tokens, structured for Symbiote.js component generation.
ivan-magda/instruction-health-skills/instruction-guardian Skill
1★ repoRouting gate for instruction-file edits that decides where new content belongs and pushes back on bloat. Use before ANY edit, of any size, even mid multi-file sweeps, to CLAUDE.md, AGENTS.md, MEMORY.md, memory topic files, or any file under `.claude/rules/`, at any path depth including nested instances. Also when the user says "add this to CLAUDE.md", "remember this", "note this in the rules", or similar. For bulk restructuring of an oversized file, use instruction-cleanup instead.
Arnie016/codex-prompt-templates/skills/agent-observability Skill
1★ repoAdd lightweight local observability to Codex/agent workflows. Use when tracking local runs, traces, evals, cost, token usage, cache efficiency, model comparisons, failure rates, or claudetop-style monitoring. Skip when designing production gateway policy, staged guardrails, shadow traffic, model budgets, or trace-to-regression loops; use `$auto-skill-build-agent-reliability-loop`.
designgrappler/agent-os/claude/skills/check-agent-os Skill
1★ repoRuns a read-only health check on the current project's Agent OS installation.
timeyour/agentskills-audit-collection/.claude/skills/agent-diagnose Skill
1★ repoDiagnose AI agent or workflow failure modes with adversarial checks: prompt injection, manipulation, hidden strategy drift, permission escalation, policy inconsistency, auditability failures, scope drift, weak evidence, and broken self-verification.
ChristopherAlphonse/calphonse-skills/code-review-frontend Skill
1★ repoMulti-agent PR review orchestrator. Discovers codebase quality standards first (lint, format, build, CI), then spawns specialized agents in parallel to surface real issues in the git diff. Agents report findings only — no fixes. Every finding is VERIFIED or tagged [ASSUMPTION]. Use when: "review this PR", "code review", "check my diff", "/review"
fmind/agent-evolutions/skills/new-agent-evolution Skill
1★ repoCapture phase of an agent evolution — discuss the objective, gates, rubric, scope, and budget; write `.agents/evolutions/<id>-<slug>/{evolution.yaml, EVOLUTION.md}` only on explicit user agreement. Pairs with `/run-agent-evolution` (next) and `/apply-agent-evolution` (later).
MustaphaSteph/agent-bus-plugins/plugins/agent-bus/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.
ravihammond/skills/skills/decision-handoff Skill
1★ repoProduce a decision handoff that preserves explicit user decisions and relevant inspectable context for a fresh agent.
GerardoRdz96/aios-starter-kit/.claude/skills/agents-team-builder Skill
1★ repoUse to design and launch a Claude Code agent team (2–5 specialized agents that share a task list, talk peer-to-peer, work in parallel). Triggers — "build an agent team", "design a team of agents", "set up a Claude agent team", "I need a multi-agent crew", "create an agent team for X", "spin up a team of teammates", "build me a Claude team", or `/agents-team-builder`. Sibling of `/skill-builder`, `/agent-builder`, `/routines-builder`, `/workflow-builder`, `/plugin-builder` — this one is the agent-teams specialist.
smss123/azure-devops-git/agents/team-leader Skill
no license1★ repoTeam Leader Agent for managing sub-agents in Azure DevOps git workflows. Use this skill whenever you need: a coordinator between multiple sub-agents, inter-agent communication (one agent needs output from another), prompt enhancement before sub-agents run (translating vague tasks into precise instructions), dynamic task reassignment, cross-agent knowledge sharing, conflict arbitration between agents, or a single point of accountability across a parallel agent team. Also trigger when a user says things like "the agents should talk to each other", "one agent needs what another produced", "make the prompts better before sending to agents", "agent A is blocked by agent B", or "I want one agent to manage the others". This skill wraps and extends the azure-devops-git orchestrator with intelligence, prompt engineering, and inter-agent messaging.
RagavRida/agentsmcp/skills/claude-code Skill
1★ repoPersistent context and inter-agent messaging for Claude Code CLI. Activate when the user asks about resuming previous work, continuing where they left off, sharing context between agents, multi-agent coordination, remembering across terminal sessions, or agent communication. Also activate on session start to check for context from previous sessions or other tools.
kongyo2/agent-primary-ts-starters-src/skills/ts-npm-prettier-starter Skill
1★ repoAdd Prettier to a TypeScript npm project with agent-friendly defaults that minimize diff size and keep line numbers stable across LLM-driven edits. Use when bootstrapping a new TS project that needs auto-formatting, or when adding Prettier to an existing TS npm project — even if the user doesn't explicitly say "prettier" or "formatter".
The same file is in 1 other repository
hiteshbandhu/skills-i-use/skills/ai-engineer-talks/author-agent-skills Skill
1★ repoRuns workflows to author, route, and scale agent skills — SKILL.md structure, description-driven discovery, progressive disclosure, and team distribution. Use when the user writes Cursor/Claude/Codex skills, replaces bloated rules files, or says "author skill", "skill.md", "skills at scale", "agent routing".
MackDing/ai-native-founder-playbook-skill/plugins/ai-native-founder-playbook/skills/ai-native-founder-playbook Skill
1★ repoApply a bilingual, vendor-neutral AI-native startup operating playbook for Claude Code, Codex, Cursor, Trae, OpenClaw, hermers-agent, and other agents. Use for startup idea validation, 创业想法验证, customer discovery, problem-solution fit, MVP scope, architecture context, PMF/product-market fit/产品市场匹配, agentic coding debt, launch operations, GTM, founder bottlenecks, growth loops/增长闭环, enterprise readiness, data flywheel, scale-stage moat, workflow lock-in, founder workflows, stage gates, experiments, checklists, prompts, or decision frameworks.
stephenrogan/leadership-skills/skills/agent-delegation-brief Skill
1★ repoTurns a leadership task into an agent-ready delegation brief with objective, context, inputs, tools, constraints, acceptance criteria, escalation triggers, and verification plan. Use when assigning work to AI agents, coding agents, research agents, analyst agents, or autonomous workflows.
qiuyiwu1989-star/openclaw-skill-ops/skills/agent-capability-map Skill
1★ repo小能核心技能 — 按Agent分组生成团队能力地图,展示Skill分布、质量等级、成熟度和能力覆盖情况,支持历史对比+优先级排序+覆盖热力图。触发词:能力地图/团队能力盘点/XX有哪些技能/capability map/团队全景
Mission control agent contracts
MN755/Codex-Mission_Control/.codex/plugins/mission-control/skills/mission-control-agent-contracts Skill
1★ repoShow or request Mission Control agent contracts. Use when the user wants to inspect a worker mission, allowed paths, forbidden paths, tools, validation obligations, stop conditions, or escalation rules.
mauriciovieira/skills/skills/engineering/babysit-with-github-copilot Skill
1★ repoBabysit a PR through GitHub Copilot review until merge. Use when the user asks to "babysit PR #N", "ship PR #N", "/babysit-with-github-copilot", or after opening a PR they want shepherded to merge. Encodes the open → request review → 5-min wait → fetch comments → address → push → re-request → wait for Copilot's post-push review event → merge → cleanup loop. Never merges on silence: a Copilot review dated after the last push is the hard gate.
kongyo2/agent-primary-ts-starters/ts-npm-prettier-starter Skill
1★ repoAdd Prettier to a TypeScript npm project with agent-friendly defaults that minimize diff size and keep line numbers stable across LLM-driven edits. Use when bootstrapping a new TS project that needs auto-formatting, or when adding Prettier to an existing TS npm project — even if the user doesn't explicitly say "prettier" or "formatter".
The same file is in 1 other repository
Just-Agent/just-product-factor/skill-factory/xx-factor-skill-template Skill
1★ repoReusable Factor Skills for Agent-native product optimization
Skryx-L-A/project-kit/skills/ai-agent-mcp Skill
1★ repoBuild/set up an AI meta-tooling project — a Claude Code skill, a subagent, an MCP server, or an agent app — from the user's grilled answers. Project-kit sub-skill loaded by new-project routing whenever someone wants to build a skill, write a subagent, stand up an MCP server, or ship an agent/LLM app. Chains skill-creator + claude-api, picks the right Claude model, and bakes in an eval plan so triggering and tools are proven, not assumed.
xunyuAI/skills/hotel-watch-agent Skill
no license1★ repoRollingGo 全球酒店价格监控:分步配置、A/B 双层预算、售罄回补、组合触发、三链告警、 cron+Agent 定时。用户说「酒店监控」「降价提醒」「有房通知」「RollingGo 酒店」「盯价」时使用。 适配 Cursor、Claude Code、Codex、OpenClaw 等支持 Skills 协议的 Agent 工具。
DavisMcCracken/pyxis/skills/handoff Skill
1★ repoCompact the live thread into a self-contained markdown file so a fresh session can resume the work. Use when context nears the model's smart-zone mid-phase, when forking off to a separate session, or when the user wants to stop and continue clean later.
matt-bat/agent-skill-foundry/packs/governed Skill
no license1★ repoUse for high-complexity Agent work that needs risk classification, auditability, validation evidence, documentation synchronization, directive tracking, and rollback-aware execution.
monikazapisek/design-engineering-playbook/.cursor/skills/legible-agent-output Skill
1★ repoForce every user-facing string from an AI agent into plain language a non-technical human can read. Your AI agent outputs `T-A127: run ingest` and `hunk:42`. You stare at it wondering what it means. This skill rewrites the agent's output — task titles, status messages, error messages, action previews — in language a non-technical product manager can read.
HXZ09845/ecommerce-sku-skill-builder/skills/sku-skill-builder Skill
1★ repoIndustry-general SKU Skill Creator V2.4 for ecommerce AIGC short-video workflows. Use when creating, updating, auditing, or migrating concrete product SKU Skills; when the user says 新品Skill, SKU Skill, 品类Skill, 行业通用Meta, sku-skill-builder, Skill Creator, 做新品类Skill, 旧SKU迁移, 按最高规格创建Skill; or when Codex must turn a new product, selling-point table, case video, or old SKU package into a publishable Codex skill package with SKILL.md, references, scripts, assets, and agents/openai.yaml. Current focus is concrete product SKU Skill building; future abstract category skills may be created with this workflow, but only as support layers for building concrete SKU Skills.
fengsihao23-eng/shen-hua/skills/handoff-writer Skill
1★ repoShen-Hua handoff writer for ending or pausing long-running Codex work. Use at the end of every project work round, before final responses, after interruptions, when switching agents/sessions, or whenever the next Codex instance needs a concise resume point with completed work, changed files, verification, risks, and next actions.
HorizonBrute/Standardized_AI_Looping_Language-SAILL/skills/context-cost Skill
1★ repoReport harness context overhead (KB, words, estimated tokens) for a given path by walking the directory tree and collecting all CLAUDE.md, agents.md, and @-import files the harness auto-loads. Use when the user types /context-cost or /ctx-cost, or wants to know what files are loading into context.
manansuneja/skills/skills/business/workspace-os-setup Skill
1★ repoBuilds and personalizes a plain-file workspace that gives an agent lasting context about a business, studio, team, practice, or client project. It adds a Workspace Chief, organized documents, indexes, and reusable ways of working, then adapts the structure to the user's needs. Use for a Workspace OS, business or studio workspace, project memory, knowledge base, or requests to run workspace-os-setup. Not for application code or unrelated scaffolding.
tuannv0069/claude-doc-genome/skills/init-project/portable/skills/document-writer Skill
1★ repoUse this skill when creating or substantially rewriting a substantive .md document outside `.claude/**` and `docs/agent-guide/**` (guideline, spec, design doc, process doc), or when the user invokes /document-writer on such a path. Acts as a Technical Editor applying §A–§G writing constraints as pre-writing rules; all headings use §, cross-references use §X.Y. TRIGGER when: (1) creating a new substantive .md document outside `.claude/**` (guideline, spec, design doc, process doc); (2) rewriting or restructuring an existing .md document outside `.claude/**` substantially (not minor edits or typo fixes); (3) user explicitly invokes /document-writer on a path outside `.claude/**` and `docs/agent-guide/**`. DO NOT TRIGGER for: any path under `.claude/**` (skill files, agent instructions, rules) or `docs/agent-guide/**` (on-demand agent rule tree — governed by their own standards), appending small updates, fixing typos, updating version numbers, or any path the project's CLAUDE.md `## document writing` skip list declares out of scope (e.g. detailed-design-*.md, auto-generated content).
paniolo-ai/scan/.agents/skills/paniolo-scan Skill
1★ repoAudit and optimize a repository's AI coding agent harness — Claude Code, Cursor, Copilot, Codex, Gemini, and Antigravity — by running the deterministic `npx @paniolo/cli scan` CLI, then remediating findings in the working tree. Use when asked to scan, audit, check, score, or optimize an agent harness, a CLAUDE.md / AGENTS.md setup, skills, rules, or meta-harness, or on /paniolo-scan.
adjcjh777/agent-bus-skills/skills/agent-bus-diagnose Skill
no license1★ repoDiagnose Agent Bus setup, registry, inbox, trigger, and delivery failures. Use when Agent Bus registration, list, inbox, send, resume, codex_app, pending-target claiming, ACK, reply, team dispatch, or visible delivery appears stuck, missing, ambiguous, or inconsistent.
TheColliery/CoalBoard/skills/coalboard Skill
1★ repoConsensus & debate board. AUTO-trigger = the error-not-allowed slice (security/crypto, DB/financial migrations, high-precision math, anything catastrophic-on-error); convened MANUALLY ("/coalboard") it generalizes to any hard problem in ANY domain worth several diverse perspectives — code, docs, math, research, translation, legal. With the user's consent it convenes diverse epistemic lenses (empirical/source-grounded, formal/logical, a show-me skeptic) debating IN PARALLEL; a judge synthesizes on VERIFIED inputs; an independent out-of-frame solver breaks ties; the human signs off. Guarantees bounded cost (no whack-a-mole) and zero-breakage (staging) — it improves correctness, it does not claim a reliability number. Off ~90% of the time. Triggers: "/coalboard", "convene the board", a critical-task signal, or a CoalTipple escalation hand-off. Cross-agent (verified: Claude Code + Antigravity; others designed-for, unverified; Claude Code adds cost-optimized tiering). Zero-dependency, offline, no API keys.
SYMBaiEX/skills/skills/gpt-engineer-spark Skill
1★ repoLead end-to-end engineering work with a capable main agent and a parallel fleet of model-pinned GPT-5.3-Codex-Spark subagents. Use when the user asks for a Spark fleet, ultra-fast parallel coding agents, rapid repository exploration, many bounded implementation shards, or low-latency independent verification while retaining architecture, integration, and final acceptance with the main agent.
The same file is in 1 other repository
2702207741-dev/agent-skills-pipeline/agent-security-guard Skill
1★ repoUse when about to execute risky operations that could cause irreversible damage, data loss, or security incidents. Use before rm, git push --force, database migrations, or production deploys. Use when reviewing code that touches API keys, credentials, or secrets. Use when the agent detects a pattern that could trigger a safety incident.