Context & AI engineering
25,991 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
Agentic workflow guard auditor
guorunjie/agentic-workflow-guard/skills/agentic-workflow-guard-auditor Skill
1★ repoUse before granting AI automation write access, reviewing agentic GitHub Actions, Bitbucket Pipelines, GitLab CI, Travis CI, Drone CI, TeamCity, Harness CI/CD, Tekton Pipelines, Argo Workflows, AWS CodeBuild, Google Cloud Build, CircleCI, Azure Pipelines, Jenkins, Buildkite, auditing n8n, Dify, Flowise, Langflow, Node-RED, Make, Pipedream, Airflow, browser automation, or checking MCP tool configs for prompt-injection and side-effect risks.
TurnkeyIsaiah/vibe-prompt Skill
1★Generate an optimized Claude Code kickoff package (spec + initial build prompt + feature-by-feature prompt sequence + workflow checklist) for building an app. Use when the user invokes /vibe-prompt or asks to "vibe code", "build an app", "help me prompt", describes an "app idea", or asks for a "coding prompt", "prompt to build", or "kickoff prompt". Turns a vague app idea into well-structured prompts grounded in Anthropic's official Claude prompting best practices, and teaches the correct build process (always starting in plan mode).
DeanThompson/agent-skills/skills/agent-context-files Skill
1★ repoCreate, review, and optimize persistent agent context files such as AGENTS.md, CLAUDE.md, .cursorrules, GEMINI.md, or repository-specific AI assistant instructions. Use when the user asks to create or improve agent instructions, project memory, coding-agent onboarding docs, or context files that should be loaded across sessions.
fmind/agent-evolutions/skills/apply-agent-evolution Skill
1★ repoApply phase of an agent evolution — diff the winner workspace against the live repo, get explicit user confirmation, copy files in, optionally re-run gates against the live tree. Writes `applied`. Requires `winner` set by `/run-agent-evolution`.
MustaphaSteph/agent-bus-plugins/claude-code/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.
krkn-s/ai-workspace/.pi/skills/skill-creator 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
gtskevin/skill-guide Skill
1★Use when the user wants to discover, inspect, compare, diagnose, or choose installed Agent Skills across Codex, Claude Code, cc-switch, or plugin directories.
nelsonwerd/prompt-pack-skill/prompt-pack Skill
1★ repoAuthor and execute "prompt packs" — a big job broken into a sequence of self-contained, independently-shippable prompts, each run in its own fresh chat so a single change never dies to a context/token limit. Also writes paste-ready handoff briefings to resume a dying chat or relay work to another tool (e.g. Codex). ALWAYS invoke when the user says any of "make a prompt pack", "turn this into prompts", "break this into phases", "prompts I can run in separate chats", "I'm running out of tokens/context", "write me a handoff", "a prompt I can paste into a new chat", "this is too big for one chat", "build plan I can hand off", or "execute Pxx / phase X from the pack". Also invoke proactively when a requested change is too large or too risky to finish safely in one session and should be decomposed into ordered, verifiable steps.
The same file is in 1 other repository
Moeblack/subagent-orchestration-skill/skills/subagent-orchestration-skill Skill
1★ repoCoordinate multiple subagents or delegated workers for parallel research, adversarial review, multi-role review, project slicing, and approval loops. Use when a user asks for multiple agents, subagents, parallel review, cluster research, multi-agent validation, or iterative blocker convergence.
wangjs-jacky/garden-skills-study/garden-skills/skills/gpt-image-2 Skill
no license1★ repo面向 GPT Image 2 的图像生成 / 编辑技能。可在 3 种环境下使用:(A) Garden 本地模式,通过 OpenAI 兼容接口直接出图并落盘;(B) Host-Native 模式,把本 Skill 当作提示词工程指引,把渲染好的 prompt 交给宿主 Agent 自带的图像工具出图;(C) Advisor 模式,宿主无任何图像工具时退化为高质量 prompt 顾问。涵盖 18 大类、80+ 个结构化模板,覆盖海报 / UI / 产品 / 信息图 / 学术图 / 技术架构图 / 漫画 / 头像 / 流程板 / 电影分镜 / IP 周边 / 编辑工作流等场景。
The same file is in 2 other repositories
unhingged/council/skills/council Skill
1★ repoRun a question, decision, or idea through an agentic 5-advisor council. Five advisor subagents argue from independent, locked perspectives in parallel — each in its own isolated context — optionally over multiple debate rounds, then a Chairman subagent renders a verdict. Use this skill ONLY when the user explicitly invokes it by name — phrases like "run this through the council", "council this", "convene the council", "use the council skill", "give this to the council", "pressure-test this with the council", or similar explicit invocations. It is heavy by design (six or more subagents per session) and should not fire on ordinary decision-shaped questions.
GerardoRdz96/aios-starter-kit/.claude/skills/agent-builder Skill
1★ repoUse when someone asks to build, create, design, optimize, or audit a Claude Code subagent (an .claude/agents/*.md file), says "make me an agent", "build a subagent", "audit my agents", or wants help deciding whether something should be an agent at all. Runs a decision gate and Discovery Interview before writing files.
andrey-learning-machines/swe-harness/plugins/swe-harness/skills/adk-agent-engine Skill
1★ repoBuild, deploy, and verify Google Agent Development Kit (ADK) agents on Vertex AI Agent Engine, including source-based packaging, custom Discovery Engine or Vertex AI Search retrieval tools, Gemini Enterprise registration preparation, and log-driven troubleshooting. Use when Codex needs to scaffold a Python ADK app, create root and sub-agents, wire custom search tools, separate local credentials from runtime environment variables, deploy to Agent Engine, inspect reasoning engine logs, or smoke-test local and remote agent behavior.
Xuepoo/agent-lx-music-skills/agent-lx-music Skill
no license1★ repoControl agent-lx-music (alx) CLI to search, play, download music, fetch lyrics/covers, and manage playlists.
cskwork/codex-goal-handoff Skill
1★Use when delegating a long-horizon, multi-step engineering task to OpenAI Codex CLI's /goal Ralph loop with durable specs, explicit human gates, and verification evidence. Triggers include "codex로 위임", "codex goal", "/goal", "ralph loop", "long-horizon task", "long-running task 자동화", "delegate to codex", and "ticket to fix to deploy 자동화".
sametbrr/prompt-architect Skill
1★Architects any user input (in any language) into a structured, domain-aware expert prompt — classifies the domain, selects the right prompting patterns, drafts a refined English prompt, runs an 8-gate self-review, and optionally executes the deliverable. Use this skill whenever the user wants to turn a rough idea or request into a professionally engineered prompt, especially when the input involves strategy, UX, product, marketing, technical writing, business analysis, or any specialized domain. Trigger this skill when the user says things like "prompt yaz", "refine my prompt", "turn this into an expert prompt", "domain analizi yap", "strateji hazırla", "plan oluştur", or provides a task description in any language and wants it elevated into a structured, actionable expert-level output. Always use this skill when the request involves multi-step domain classification + prompt generation + optional execution.
ppezzull/use-dockploy-mcp/skills/run-dokploy-mcp Skill
1★ repoDeploy, manage, and monitor services on Dokploy via MCP tools. Use when deploying apps, checking logs, managing domains, updating environment variables, or inspecting container status on the Dokploy host.
nzl153/memory-system Skill
no license1★10层记忆系统 — 日记、搜索、图联、会话索引、能力自检。
kongyo2/agent-primary-ts-starters-src/skills/ts-npm-oxlint-starter Skill
1★ repoAdd Oxlint to a TypeScript npm project with agent-friendly defaults that focus error-level diagnostics on real bugs, silence stylistic noise, and emit output in oxlint's `agent` format for downstream LLM consumption. Use when bootstrapping a new TS project that needs linting, or when adding Oxlint to an existing TS npm project — even if the user doesn't explicitly say "oxlint" or "linter".
The same file is in 1 other repository
zrosenbauer/skills/skills/code-reviewer Skill
1★ repoThis skill should be used when the user wants to review code, audit a diff, get a second opinion on changes, or run an adversarial review of files in the current working tree. Common triggers include "review this code", "audit this diff", "find issues in", "second opinion on this", "harsh review of", "adversarial review", and "security review of". Picks one or more reviewer personas (adversarial, security, architecture, performance). Reviews local files, `git diff`, or `git diff --staged` only — does not fetch external content. Runs in one of four modes: single-agent (one persona in the current agent), cross-model handoff (independent second opinion via another local AI CLI, with secret-shield preflight + prompt-shield wrap), multi-bg-agent (one persona per parallel background subagent), or agent-team (Claude Code Teams or equivalent on supporting agents). Skip when the user wants formatting fixes (use a linter) or refactoring patterns (use ts-best-practices or ts-best-practices-functional).
MackDing/ai-native-founder-playbook-skill/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.
sjh6229/agent-continuity/skills/agent-continuity Skill
1★ repoUse when initializing, resuming, continuing, handing off, or closing a long-running AI coding project across sessions, especially for initialize continuity notes, set up handoff notes, create docs, resume work, continue later, handoff, next session, long-running project, session state, continuity notes, docs/START_HERE.md, docs/CURRENT_STATE.md, docs/PROGRESS.md, docs/DECISIONS.md, or the legacy keyword longwork.
0SxD/candidacy-checker-eval Skill
1★Use this skill when a candidate wants to evaluate whether they are ready to apply for a specific role before submitting. The skill runs a two-layer protocol: (1) a Socratic profile discovery loop that extracts the target role and candidate state at 95%+ confidence across three independent lenses (Logos, Pathos, Ethos), then (2) a flat boolean checklist rubric covering resume, LinkedIn, portfolio, writing samples, references, application materials, JD requirements, interview prep, and honesty gates. Each rubric row outputs PASS, FAIL, UNKNOWN, or NOT_APPLICABLE with a remediation note. Trigger when a candidate says they are about to apply for a role and wants an honest gap assessment, or when an AI assistant is preparing application materials and needs to audit readiness first.
stephenrogan/leadership-skills/skills/agent-command-center Skill
1★ repoCreates a command-center view for managing multiple AI agents and human-agent workflows, including active work, owners, state, blockers, quality signals, risk level, approvals, and next decisions. Use when a leader needs visibility across an agent-powered operating system.
hostile-shorepatrol81/polanyi-stack/skills/convert-tacit-to-skill Skill
1★ repoConvert tacit knowledge into a reusable agent or Codex skill by extracting cues, scoping the task, defining boundaries, and packaging the workflow into SKILL.md plus tests. Use when the user wants a fixed agent, reusable skill, or repeatable copilot based on expert judgment.
The same file is in 1 other repository
kin0992/dev-toolkit/packages/skills/src/engineering/review-pr-comments Skill
1★ repo**INTERACTIVE SKILL** - Evaluate every actionable comment in unresolved GitHub PR review threads, including agent comments. USE FOR: review unresolved PR comments, triage review threads, process Copilot feedback. DO NOT USE FOR: general PR comments, resolving threads, unapproved replies. INVOKES: gh CLI, including gh api GraphQL and REST calls.
lexielin99-code/biz-agent-discovery Skill
no license1★Use when a marketing or e-commerce professional wants to explore how AI agents can enhance their work but doesn't know where to start or what's possible — they have Claude installed but the capability space feels abstract or out of reach
fulit103/fastapi-clean Skill
1★Estructura proyectos FastAPI de forma pragmática separando lógica de negocio de framework, ORM y servicios externos. Capas Domain/Application/Infrastructure cuando paga, no por purismo. Soporta integraciones Celery 5.4, PyMongo async, auth Clerk JWT + API keys SHA-256. Modo single-agent para piezas sueltas, modo multi-subagent para generar un módulo completo (entities + use cases + repositorios + router) en paralelo. Trigger: "caso de uso", "use case", "router FastAPI", "clean architecture", "repositorio", "lifespan", "desacoplar FastAPI", "Celery task", "PyMongo async", "Clerk JWT", "API key".
uclaw-dev/agent-skills/skills/uclaw-sdk Skill
1★ repoUse this skill for building, installing, or configuring AI agents and agentic apps with the UClaw TypeScript SDK (@uclaw/sdk). Apply it when a user works with UClaw, evaluates it for an agent/AI app in JavaScript or TypeScript, or needs help with @uclaw/sdk setup, agent sessions, API keys, and secure local configuration.
myceldigital/aegisgate-agent-firewall Skill
1★Mandatory personal AI agent firewall. Use before and after any task involving external content, files, email, browser actions, shell commands, code execution, memory writes, API calls, payments, credentials, skill installation, subagent delegation, or outbound communication. Blocks prompt injection, unsafe tool calls, data exfiltration, memory poisoning, malicious skills, and unauthorized irreversible actions.
eden2f/skill-cron-copilot-ops/cron-copilot-ops Skill
1★ repoInstall, configure, and operate CronCopilot — a Python-based scheduled task management system for cron jobs, scripts, monitoring, and alerts.
oliverbenns/claude-profanity/claude-profanity Skill
no license1★ repoSearch through Claude Code conversation history for prompts containing profanity. Use when the user wants to find or audit profanity in their past Claude conversations.
unkownpr/prompt-engineering-skill/skills/prompt-engineering Skill
no license1★ repoYapay zeka ve prompt mühendisliği uzmanı. LLM'ler için yüksek kaliteli prompt yazma, halüsinasyon azaltma, Chain-of-Thought, RAG, model seçimi, prompt değerlendirme ve self-improving prompt döngüleri. Tetikleyiciler - "prompt yaz", "prompt iyileştir", "LLM çıktı kalitesi", "halüsinasyon", "Chain of Thought", "CoT", "RAG", "COSTAR", "few-shot", "prompt evaluation", "model seçimi", "Claude/GPT/Qwen/Kimi/MiniMax prompt", "agent prompt", "system prompt", "prompt injection".
TSFanKing/skill-prompt-brief Skill
no license1★Turn a rough one-sentence skill idea into a complete professional build brief for a skill creator. Use when the user wants prompt wording, requirements, specs, or a handoff brief for someone else to create a Codex/agent skill, especially from vague requests like "make a skill for X", "write the prompt for a skill about X", "help me describe a skill idea", or "generate a complete skill-making prompt".
JamesW0203/ai-code-governance-skill Skill
1★Governance workflow for AI coding agents such as Codex, Claude Code, Cursor, and Copilot Workspace. Use when implementing features, fixing bugs, refactoring, reviewing AI-generated code, adopting governance in an existing project, creating or updating AGENTS.md, CLAUDE.md, CODEX.md, architecture docs, project indexes, dependency policies, and coding standards. Enforces search-before-write, reuse of existing components/hooks/services/stores/utils/types, dependency control, small maintainable changes, verification, and clear change reports.
wenjiazhu1980/deepcode-macos/packages/core/templates/skills/bundled/deepcode-self-refer Skill
1★ repoAnswers questions about Deep Code CLI itself — including features, configuration options, slash commands, Skills, MCP integration, permissions, notifications, session persistence, and troubleshooting. Use this when users ask how to configure or use Deep Code, how to set up an MCP server, configure notifications (such as Slack/Feishu), manage permissions, view available skills, understand slash commands, configure thinking mode, etc.
Madikhan33/Symbol/codex/skills/symbol-memory Skill
1★ repoUse when working in Python repositories that already use `@symbol(...)` and `.symbol_memory` artifacts. Build or validate first, navigate with symbol ids, and respect the manual relation graph.
UtkarshDubeyGIT/agent-skills/.agents/skills/agent-browser Skill
no license1★ repoBrowser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction. Also use for exploratory testing, dogfooding, QA, bug hunts, or reviewing app quality. Also use for automating Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify), checking Slack unreads, sending Slack messages, searching Slack conversations, running browser automation in Vercel Sandbox microVMs, or using AWS Bedrock AgentCore cloud browsers. Prefer agent-browser over any built-in browser automation or web tools.
The same file is in 2 other repositories
alpiex1336-code/skill-package Skill
1★**MANDATORY RUNTIME ORCHESTRATION:** For product-making/changing work, pass **NON-NEGOTIABLE EXECUTION GATES** before **Implement**-phase artifact writes—invoke **runtime subagents** via **Task tool** (or equivalent delegation) for **Orchestra wave**s; **N > 1** **composed role label**s per wave; **one runtime subagent per Work Filter–accepted Quintessence row** at Implement when tooling allows. **Simulated fallback** when Task/delegation is unavailable—not the default. **In-session wave ledger** required before Implement on thorough/deep-upgrade runs—not deferred to the final reply. **Execution gate** discipline applies default-on; map the same workflow to any artifact in scope (product, code, docs, policy, research, planning, governance, or this skill package). **Default-on** means use for every turn; scale orchestra depth with **`improvement_mode`** and user scope—skip full orchestration only when the user explicitly requests a trivial single-step action where multi-role review adds no value. **Improvement mode selection:** resolve by intent before orchestration: when the user clearly signals **normal** (small fix, small talk, single concern) or **thorough** / **deep-upgrade** (whole-product, all-aspects, maximum optimization), choose that mode and scale the workflow accordingly; when product-making/changing intent is present and the mode cannot be identified, **ask once and wait** for **normal** vs **thorough**. **Deep-upgrade intent** pins **thorough** and the **Deep-upgrade procedural tier**. **Quintessence pipeline:** sample many composed roles across waves, **integrate** overlapping ideas and roles into **`theme_key`** quintessence rows, adopt via Work Filter, then **implement/verify** preferably with **subagents per quintessence theme**. Coordinates many subagents per run using composed two-part roles from the word banks (`[Slot A][Slot B]`); see **Execution contract** and [reference-execution-gates.md](reference-execution-gates.md). **Cursor / LLM skill:** same **Stochastic Role Orchestra** pattern for any assistant that reads this file. Expects **universal aspect-coverage** discipline—build a **target coverage set** per engagement (see **Lexicon**), then **bias** role draws toward **uncovered** aspect classes. Uses **word-bank priority tiers (P0 / P1 / P2)** so **`improvement_mode` = thorough** whole-product runs have a **session floor** of **P0 families** (see [reference-word-banks.md](reference-word-banks.md)). **`normal`** biases **incremental modification**; **`thorough`** authorizes **material reshaping** while preserving **north-star invariant**. Requires epistemic honesty (no invented context; anti-sycophancy). **Outward closure:** after honest saturation (**Workflow §8**), terminal replies stay decisive. **Voice:** professional register; avoid LLM-default “AI voice.” **Skill-targeting requests** use **Recursive self-application**. Triggers include “essential fixes only,” “surgical,” “full product pass,” “**thorough** pass,” “multi-agent review,” “orchestrated review,” **mature product iteration**, greenfield/brownfield maturity passes. **Supersedes** informal fixed job-title maturity loops by mapping concerns onto **composed roles** + depth references—stricter anti–single-voice discipline.
roebi/sw-dev-agent-framework/skills/apply-tdd-loop-en Skill
no license1★ repoApplies the Test-Driven Development (TDD) cycle as an agentic loop of RED -> GREEN -> REFACTOR iterations until all requirements are covered or all existing code paths are tested. Use when a user or agent wants to write code strictly test-first, add missing tests to existing code, or drive a full TDD session for a module, class, or function. Activate for trigger phrases like: "apply TDD", "write this test-first", "do a TDD loop", "TDD cycle for", "drive TDD on", "red green refactor", or "add tests then implement".
apertis-ai/apertis-skills/skills/apertis-api Skill
no license1★ repoUse Apertis API to access 500+ AI models with OpenAI-compatible SDK. Covers authentication, endpoints, popular model families, web search (:web suffix), the Vercel AI SDK provider, and MCP server setup.
MonkeyUI-dev/shadcn-component-lock Skill
1★Generates and maintains a `shadcn-component-lock.md` file at the project root that tells Vibe Coding agents which files in the shadcn UI directory are stock shadcn/ui primitives — reused across the project and NOT to be edited directly. Use this skill whenever the user runs `npx shadcn@latest add`, `init`, or `apply`; whenever new files appear under the project's shadcn `aliases.ui` directory; whenever the user asks to "lock", "freeze", "pin", "regenerate", "refresh", or "update" the shadcn component lockfile; or whenever a Vibe Coding agent (e.g. Cursor, Copilot, Codex, Claude Code) is about to modify a file under `components/ui` and needs to know whether it is safe to touch. Coordinates with the official `shadcn` skill from shadcn-ui/ui.
ymy88/agent-coding-conventions Skill
1★Conventions for writing code and using developer tools reliably as an AI coding agent. Use when judging how far to trust a code-intelligence / static-analysis / indexing tool's output (e.g. codegraph) — especially negative claims like "no callers", "no covering tests", "impact is only these files"; or when a design choice (factory vs public constructor, DI, reflection, dynamic dispatch) affects how legible the code is to an agent. Consult BEFORE trusting a tool's claim as fact, and BEFORE changing a design to satisfy a tool.
gzb1128/skill-forge/plugins/agent-docs/skills/bootstrap-agent-docs Skill
1★ repoSkill Forge: Claude Code plugin marketplace for agent harness docs, code quality workflows, and OpenCode customization.
richard-kim-79/popup-skills/popup Skill
1★ repoCreate instant shareable web pages on popup2026.com — no login, no PIN, 30-second setup. Use when the user wants to make a "팝업 페이지", "공유 링크", "랜딩 페이지", "이벤트 안내", "링크 모음", host a raw HTML file, or host a PDF. Tools are reached via the Remote MCP server at https://popup2026.com/api/mcp.
jakebrownscombe/science-lab-AI-framework/knowledge_base Skill
1★ repoA customizable framework for using LLMs in scientific workflows.
ayi-ai/vibe-coding-playbook Skill
1★A lightweight engineering workflow for turning an idea into a real, shippable product with an AI coding agent. Core = Karpathy's four coding disciplines (think before coding / simplicity first / surgical changes / goal-driven; battle-tested), plus an Explore→Plan→Build→Polish→Ship process, task tiers (skip the process for small changes, full process only for shipped products), and the "AI is an amplifier" mindset (good habits ×10, bad habits ×100; controlling complexity is still the human's job). For everyone: terse for engineers, explained for newcomers, no jargon dumps. Ships with a paste-in kickoff prompt, a drop-in .cursor/rules discipline file, and PRD/architecture/status doc templates. Use when building an app/product/website/tool from scratch with Cursor/Codex/Claude Code, when asked to set rules before coding, when worried about code rotting into a mess or the AI changing unrelated code on its own, when asked how to do "vibe coding", whether to write a PRD/architecture, how to start a new project, or when "amplifier / spaghetti code / minimal loop / surgical change / complexity / task tier" come up.
Moonweave-Research/groundcheck/arxiv-monitor Skill
1★ repoFetches new arXiv papers matching a research watchlist, scores relevance with LLM, and delivers a filtered daily digest. Invoke when the user wants to check new arXiv papers on a topic, set up a daily paper monitoring routine, asks 'what's new on arXiv about X?', or wants to scan recent preprints in their field. Also use as a Claude Code Routine template for scheduled daily briefings. Do NOT invoke for finding a specific known paper (use ref-verify) or for general web research.
kongyo2/agent-primary-ts-starters/ts-npm-oxlint-starter Skill
1★ repoAdd Oxlint to a TypeScript npm project with agent-friendly defaults that focus error-level diagnostics on real bugs, silence stylistic noise, and emit output in oxlint's `agent` format for downstream LLM consumption. Use when bootstrapping a new TS project that needs linting, or when adding Oxlint to an existing TS npm project — even if the user doesn't explicitly say "oxlint" or "linter".
The same file is in 1 other repository
findscripter/everything-skills/00-meta/agent-self-reflection Skill
no license1★ repo当对话陷入细节、用户喊"退一步/缩放视角/我们是不是想多了"或长时间深挖实现却无战略校验时使用;做对话的五维复盘(宏观视角/缺口分析/反思追问/偏见检查/上下文对齐),产出"流水式无小标题"评估并以 继续/转向/暂停提问 三选一的明确方向收尾;不适用于例行执行、需要写报告式结构化输出、或对话过短无可复盘上下文的场景;触发词:反思、退一步、缩放视角
xunyuAI/skills/flight-monitor-agent Skill
no license1★ repo机票实时查价 Skill:RollingGo 查价、精简/全量穷举、暖色 HTML 分页报告。 用户说「查机票」「穷举搜索」「RollingGo 查价」「国庆特价」「开口程」「东南亚/日本/菲律宾机票」时使用。 适配 Cursor、Claude Code、Codex、OpenClaw 等支持 Skills 协议的 Agent 工具。
Timur00Kh/cmux-agent-control-skill/skills/agent-control Skill
1★ repoRun and orchestrate multiple Pi AI agents inside cmux on macOS. Covers pane creation, non-disruptive layout management, interactive Pi sub-agent spawning, sending task prompts, reading screens, and monitoring worker sessions.
Xtqm/skills/skills/vercel-eve-framework Skill
1★ repoExpert guidance for building AI agents with Vercel's Eve framework — an open-source, filesystem-first, TypeScript-native agent framework (public beta since June 17, 2026). EXPLICIT triggers — activate when user mentions: "eve agent", "vercel eve", "eve dev", "eve framework", "npx eve@latest", agent.ts, agent/instructions.md, agent/tools/, agent/skills/, agent/channels/, agent/connections/, agent/schedules/, agent/subagents/, "defineTool", "defineAgent", "defineEval", "defineMcpClientConnection", "eve eval", "eve channels add", Vercel Sandbox in agent context, Vercel Workflows in agent context, AI Gateway in agent context. SEMANTIC triggers — activate even without "Eve" keyword when user describes: "durable AI agent on Vercel", "long-running AI agent", "agent that survives crashes or deploys", "workflow-backed agent", "checkpointed agent sessions", "agent that runs for hours / days / weeks", "agent with human-in-the-loop approvals", "filesystem-driven agent framework", "Vercel agent framework", "agent directory structure", "agent that deploys to Vercel", "agent with sandboxed code execution", "agent with isolated compute". ACTION triggers — activate when the user wants to: scaffold a new agent project on Vercel, add channels (Slack, Discord, Teams, GitHub) to an existing agent, add approval gates to agent tools, connect an agent to MCP servers, set up agent observability with OpenTelemetry, test agent behaviour with scored evals, migrate from another agent framework to Eve. This skill covers the complete Eve surface. For deep topics it routes to focused reference files, each loaded only when relevant, reducing context overhead by ~70%.
KunalKumarkkr01/design-system-extractor/skills/design-system-extractor Skill
1★ repoReverse-engineer the design system of any live website into a single, implementation-ready reference (a design.md plus annotated screenshots) that another LLM or developer can follow to faithfully recreate the look and feel. Use this whenever the user wants to capture, document, extract, clone, mirror, copy, or reproduce the design / UI / styling / "vibe" / look-and-feel of an existing site or web app — for example "document the design system of stripe.com", "make a style guide from this URL", "I want my app to look like X, extract their design tokens", or "explore this site and tell me how it's built visually". Works on any reachable URL. Requires the Chrome DevTools MCP server to drive a real browser; the skill is non-functional without it.
made-on-weekends/vibe-code-ninja/project-ninja Skill
no license1★ repoMaintain alignment between code and project documentation across AGENTS.md, CLAUDE.md, GEMINI.md, README.md, and the docs/ folder (PRODUCT, ARCHITECTURE, API, DATABASE, SCHEMA, DESIGN, BRAND, UX, COMPONENTS, DECISIONS, SECURITY, TESTING). Stack-agnostic — works for any language or framework. Use this skill at the start of any non-trivial coding session to load ONLY the relevant docs (token-efficient routing), before adding or removing features (to prevent scope drift and accidental deletions), and after changes are merged (to update only the docs that actually need updating). Trigger this skill whenever the user mentions align, drift, scope, documentation, project setup, init, update docs, refactor, or any feature add or remove — even if they do not explicitly ask for documentation work. Also trigger on phrases like "is this in scope", "did we decide", "what's the convention", or any time the user is about to delete code.
Michaelihc/claude-code-cli-codex-skill/claude-code-cli Skill
1★ repoDelegate work from Codex to the local Claude Code CLI with low-waste, current command patterns. Use when Codex needs to call `claude`, choose Claude Code model or effort settings, use `ultrathink` or `ultracode`, run print-mode JSON calls, resume or continue Claude sessions across turns, manage background Claude agents or workflows, cap spend with `--max-budget-usd`, restrict Claude tools, or compare Claude output against Codex work.
matt-bat/agent-skill-foundry/packs/developer Skill
no license1★ repoUse for general Agent engineering work that needs a compact operating policy for clarification, sequencing, safe command execution, regression control, documentation upkeep, quality review, frontend judgment, and the persistent --quizme clarification mode.
lzehrung/review-until-clean-odw/skills/review-until-clean Skill
no license1★ repoPortable pre-PR review-until-clean workflow and agent skill
udohjeremiah/skills/.agents/skills/skill-creator 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
jesse-black/skills/skills/cli-as-harness Skill
1★ repoAuthor CLI surfaces as agent-facing documentation, especially `--help` and actionable error messages. Use with any language or parser design when adding or changing a CLI interface, improving help text or validation errors, moving command reference out of AGENTS.md / CLAUDE.md / a README / a skill and into CLI surfaces, or making errors suggest the corrected command line or tool call.