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
zning1994/docs-organization Skill
2★Organize project documentation by size, audience, and freshness. Trigger when setting up docs for a new project, restructuring existing documentation, slimming a bloated CLAUDE.md/AGENTS.md, deciding where a doc should live, or on doc best practices questions. Also trigger when CLAUDE.md exceeds ~250 lines or docs are duplicated across files. Triggers on: organize docs, reorganize documentation, slim CLAUDE.md, AGENTS.md too long, where to put docs, docs structure, doc best practices, set up documentation, docs are messy, flatten docs, doc migration.
xmqywx/claude-code-skills/skills/agent-sdk-builder Skill
no license2★ repoBuild production AI agents using Anthropic's Claude Agent SDK (TypeScript or Python). Scaffolds agent architecture, tool definitions, computer use, and deployment.
ks6088ts-labs/skills/skills/agents-md-creator Skill
2★ repoAIコーディングエージェント向けの指示書「AGENTS.md」を作成するスキル。プロジェクトにAIエージェントが作業するための文脈と指示を集約するファイルを作成したい場合に使用します。「AGENTS.mdを作成」「AIエージェント用の指示書を作る」「エージェント向けREADMEを作成」などのリクエストでトリガーします。OpenAI Codex、Claude Code、GitHub Copilot、Cursorなど、複数のAIエージェントで共通利用できるオープンな標準フォーマットです。
bcastelino/agent-skills-kit/skills/ai-agent-development Skill
2★ repoAI agent development workflow for building autonomous agents, multi-agent systems, and agent orchestration with CrewAI, LangGraph, and custom agents. Use when designing or implementing AI agents.
Igapyon agent state management
igapyon/igapyon-agent-skills/skills/igapyon-agent-state-management Skill
2★ repoUse only when the user explicitly names igapyon-agent-state-management or uses an igapyon-prefixed trigger phrase such as igapyon 状態管理, igapyon agent 状態管理, igapyon 作業状態, igapyon 作業再開, igapyon goal, igapyon todo, igapyon handoff, or igapyon GOAL TODO DECISIONS HANDOFF. Do not trigger for generic TODO.md maintenance, ordinary project planning, vague handoff discussion, ordinary work resumption, broad Context Engineering talk, generic build/test failures, or even the GOAL.md + TODO.md + DECISIONS.md + HANDOFF.md workflow unless the user also says igapyon or explicitly names this skill. When active, also supports repository-local Harness Operations Decisions in DECISIONS.md for reusable build/test/package/comparison/roundtrip execution decisions.
siegerts/agent-transcript/skills/agent-transcript Skill
2★ repoFind and export agent chat session transcripts. Use when the user asks to export, save, find, or share a session transcript from any coding agent (Claude Code, Codex, Kiro, Cursor, Gemini CLI, etc.), wants to create an image of their conversation, or mentions agentpng.
darshitpp/x-agent/claude Skill
2★ repoDelegate tasks or get a second opinion from Claude Code CLI. Infers mode (delegation vs validation) from prompt context. Detects available models at runtime via the CLI. Use when cross-validating code, plans, or architecture with Claude Code, or offloading work to it. Do not use when already running inside Claude Code or for tasks that do not benefit from cross-model input.
rodrigolagodev/skill-creator Skill
2★Expert guidance for creating, structuring, and refining Agent Skills following the open standard. Use when working with skill files, authoring new skills, improving existing skills, or understanding skill architecture and best practices. Use PROACTIVELY when user mentions creating skills, structuring documentation, or building reusable agent capabilities.
CodeAlive-AI/exa-skills/exa-code-context Skill
no license2★ repoFind code with Exa — API usage examples, error fixes, library docs, and GitHub repositories. Use when you need a working code snippet, to resolve an error, find an open-source implementation, or pull up official docs for a language/framework. Runs a local script against the Exa API; no MCP server required.
BreetyGreen/Myco/skills/multi-agent-skill-sharing Skill
2★ repoInstall a skill once and make it usable across multiple AI coding agents (Claude Code, Codex CLI, Cursor, Gemini CLI, Cline, etc.) that work on the same repository. Use this when the user wants a skill / ruleset / prompt to be shared and switchable across several AI products, asks "why can only one tool use this skill", or wants cross-tool consistency. It maps each agent to its correct skill-reading directory, installs into the repo so the skill travels with Git, and documents the per-tool invocation syntax (which differs — e.g. Codex uses `$skill-name` or `/skills`, NOT gstack-style `/design` commands).
coreline-ai/codex-resstyle-skill/response-style-sync Skill
2★ repoCreate, refine, and sync project response style, agent execution rules, and LLM/proxy usage policy into AGENT.md and CLAUDE.md from reusable JSON profiles. Use when the user wants to standardize Korean-first answers, force dev-plan-first workflows, require clarification before risky decisions, block default cloud API usage, prefer proxies such as multi_model_tui, or keep AGENT.md / CLAUDE.md aligned with a shared project rule set across repositories.
cyzlmh/agentype Skill
2★Run the Agentype workflow for local AI-agent usage analysis: collect and cache deterministic JSON, infer a persona/archetype from aggregate usage signals, then render a terminal summary or PNG poster. Supports Claude Code, Codex, OpenCode, pi-agent, Gemini CLI, OpenClaw, Nanobot, and configured Nanobot-compatible roots. Use when the user asks to understand their agent usage, AI workflow, token footprint, preferred agents/models/projects, or "agentype".
dautovri/cloud-cost-agent/skills/cloud-cost-agent Skill
2★ repoUse when analyzing cloud spend, running a cost audit, or finding savings across AWS, Google Cloud, and Azure — rightsizing, idle resources, and commitment/reservation recommendations, per-provider or cross-cloud. Uses only native CLIs and recommendation services (Cost Explorer/Compute Optimizer/Cost Optimization Hub, GCP Recommender, Azure Advisor + Cost Management) to produce prioritized savings reports.
lukehalley/introspect/skills/introspect Skill
2★ repoReview the user's recent AI coding threads across Claude Code, OpenAI Codex, and Cursor to learn how THEY prefer to work — the preferences and corrections they keep repeating — and to spot places where a missing rule or script made the agent repeat work, then propose and (on approval) write those updates back into each agent's config: CLAUDE.md / memory / skills for Claude, AGENTS.md for Codex, and .cursor/rules for Cursor. So the setup compounds and the user stops giving the same instruction twice. Use this whenever the user wants to review/introspect their sessions, learn from recent threads, run a "retro" on their day/week, "close the loop" across Claude and Codex and Cursor, figure out where the agent repeated the same mistake, turn corrections into rules, reduce repeated prompting, or self-improve their config from actual usage — even if they don't say the word "introspect". Also the target for a daily or every-N-sessions self-improvement routine. On first use, or on "set up introspect" / "configure introspect", run the interactive Setup to tailor agents, window, safety, and schedule.
ZHOUCOOKIE/prompt-engineering-skill/skill/en Skill
no license2★ repoWrites, rewrites, reviews, and audits prompts and system prompts using Anthropic's current official guidance, and selects the right model and effort level for the task before drafting. Use when the user asks to write a prompt, improve or fix a prompt, design a system prompt, or check a prompt before running a task — including phrasings like "write me a prompt for", "make this prompt better", "design a system prompt", "check my prompt before I run this", "turn these requirements into a prompt". Applies whether the resulting prompt will be pasted into Claude Code for a coding task or run in a desktop chat or Cowork session for a non-coding task. Also use when the user pastes a prompt and asks why it is not working, when they are authoring an agent system prompt or CLAUDE.md, or when they ask which model or effort level to use. Several widely repeated prompt techniques are now obsolete or rejected by the API, so consult this skill even when prompting from memory feels sufficient.
abhishekgahlot2/agent-optimization/skills/agent-optimization Skill
2★ repoUse when optimizing, reviewing, or debugging LLM agents for latency, cost, reliability, delivery quality, benchmark integrity, prompt bloat, retry loops, tool-call waste, fast paths, or eval regressions. Apply the Agent Optimization Prompt: measure honestly, use real traces, exhaust deterministic levers before model/prompt tweaks, and verify on the real serving path.
coroboros/skillward/skills/skillward Skill
no license2★ repoVet an untrusted agent skill, plugin, or MCP server before it is installed or trusted. Runs the skillward CLI — the complete deterministic scanner ensemble, offline, fused into one verdict — then triages the report into a plain-language risk read with an install / don't-install / remediate call. Use this whenever someone is about to add or run third-party agent code (npx skills add, a marketplace or plugin, an https Git URL, a directory of skills, or a local folder), even when they only ask "is this safe?" rather than naming a scan. Trigger phrases include scan this skill, vet before install, audit this agent skill, is this skill or MCP server safe, check this repo before I trust it.
KevinLuo1/kw-okf-memory-skill/skills/kw-okf-memory Skill
2★ repoBuild and maintain a structured Obsidian-friendly Markdown memory wiki for Codex using a Karpathy-style wiki, an opinionated OKF-inspired KW-OKF profile, rebuildable categories.json and graph.json indexes, safe structure evolution, staged AI writes, and optional OKF v0.1 export. Use when Codex should save, update, retrieve, synthesize, search, stage, commit, audit, maintain, or export durable project memory; prevent messy AI-written knowledge bases; check duplicates, conflicts, stale notes, broken links, associations, and vault structure; process memory assets; or open committed notes in Obsidian for human review.
HenrikBrehm/prompt-refiner-skill/skills/prompt-refiner Skill
2★ repoLints an existing user-supplied prompt for prompt-engineering defects and prompt-injection risks. Produces a diagnostic report with rule IDs, severity, line/column, quoted evidence, rationale, and optional JSON output. Does not rewrite prompts or impose prompt frameworks.
Sambhram1/Ghostpatch-/skills/ghostpatch Skill
2★ repoUse Ghostpatch as an agent-native open-source contribution plugin for Codex, Claude Code, and other coding agents. Trigger when a user wants the agent to find repositories, discover GitHub issues, qualify candidates, solve issues locally, review diffs/tests, and raise approval-gated pull requests using the Ghostpatch workflow.
felixfinal/agent-dsviewer-logic-analyzer Skill
no license2★Top-level agent for USB logic analyzer work using the verified dslogic-cli/libsigrok4DSL native backend. Use for DSLogic/DSView environment setup, device connection, capture (stream+buffer), protocol decode (UART/SPI/I2C), and data interpretation. Use the dsview-logic MCP for hardware operations.
ethanolivertroy/mischief-managed Skill
2★Use when finishing a Pi, Codex, Cursor, Claude, or other coding-agent session and wanting to save a lean Markdown session note into your Obsidian vault with decisions, code changes, validation, caveats, and follow-ups.
Beichen-H/Loopower/skills/prompt-to-loop-engineering Skill
2★ repoUse when a natural-language task must be converted into a role-neutral, statically validated one-shot plan, workflow, or agent-loop specification before any user-task execution begins.
Jared13PG/ai-engineering-workflow/.agents/skills/ai-engineering-workflow Skill
2★ repoUse this skill for complex coding, coursework, data analysis, notebook, research, or multi-file projects that need structured issue creation, context analysis, task breakdown, atomic execution, verification, status tracking, and final closure.
hannsxpeter/godplans/skills/godplans Skill
2★ repoProduce an audit-aware, agent-executable master plan (PLAN.mdx) for a software project before application code is written. One command runs discovery, forces hard-to-reverse decisions, and plans product, architecture, roadmap, stack, repo, build, deploy, observability, launch, security, code quality, style genome, database, LLM integration, SEO, UI, UX, and agent memory upfront. After-the-fact audit checks become plan-time acceptance criteria, and a self-contained validator enforces task structure and approval state. Use when the user says: plan this project, godplans, master plan, plan everything upfront, idea to plan, plan before code, audit-aware plan, replan, or starts a greenfield project or major feature. Refuses plan theater (sections filled, decisions absent), vague tasks without verification, unsupported quality guarantees, and projects whose core purpose violates the Anthropic Usage Policy.
bogdanbaciu21/skills/agent-coordination Skill
2★ repoCoordinate a deliberation across multiple AI agents (Claude, Gemini, Codex, Cursor, Antigravity, etc.) using a shared markdown log. Teaches Claude to read the full log before responding, append entries with the locked preface format, follow the phase model (Research → Converge → Stabilize → Action-List → Approval → Execute → Closeout), enforce the approval gate, and hand off cleanly to the next named agent. Also scaffolds a new coordination session on request. Triggers on "agent coordination", "coordination session", "multi-agent", "round-robin", "files-received", or any mention of `agent-research/agent-coordination.md` or `agent-research/files-received/`. SKIP for generic single-model multi-turn work — this skill is only for sessions where the human is brokering between multiple distinct AI agents through a shared file.
jukrap/ai-agent-playbook/skills/ai-harness/agent-orchestration-handoff Skill
2★ repoUse when splitting work across agents, subagents, workers, review passes, or long-running handoffs.
ainova-systems/intelligence-dev-packs/packs/core/skills/dev-handoff Skill
2★ repoCompact the session into a self-contained continuation prompt for a fresh agent session
adtn0810/truestack/skills/truestack-agent-coordination Skill
2★ repoCoordinate multiple agents and multiple sessions on the same repo without overwriting or colliding with each other. Use whenever work is big enough to split across parallel agents, when a research or planning task is explicitly split across parallel sub-agents (the research content itself belongs to truestack-deep-research), or when continuing work another session started. Covers task decomposition, worktree isolation, a shared task ledger, and the merge protocol.
Imtylerrrrrr/TK/en Skill
2★ repoUse when starting or joining a team project or competition repository and collaboration rules need to be set up — commit/PR/branch conventions, agent context files, GitHub merge settings. Triggers — new hackathon/competition repo, "collaboration rules", "commit convention", "PR rules", "set up team repo", team develops with coding agents (Claude Code, Codex, Cursor, Gemini CLI).
Bmob cloud function development
bmob/agent-skills/skills/bmob-cloud-function-development Skill
2★ repoWrite, upload, and verify Bmob server-side cloud functions using the Bmob MCP server when available. Use when the user asks to 编写云函数, 写云函数, 上传云函数, 部署云函数, 发布云函数, 验证云函数结果, or mentions `function onRequest(request, response, modules)`.
m-binimran/dev-pack/meta/claude-pack-builder Skill
2★ repoDesign, build, verify, and publish a curated Claude Code pack — a reusable collection of rules, skills, hooks, loops, and review agents — to GitHub. Trigger this skill whenever the user wants to "create a Claude Code pack", "build a rules/hooks/skills collection", "make a guardrail pack", "package my hooks/skills/agents", "publish a Claude Code config to GitHub", or assemble enforced standards for a stack (e.g. Next.js + Supabase, Django, etc.). This skill is the methodology that produced github.com/m-binimran/dev-pack; reuse it instead of re-deriving the process. Run it BEFORE writing any pack files — it sequences research, clarification, building, verification, and safe publishing.
shandar/memory-map/skills/memory-map Skill
2★ repoTwo-tier project memory system. Maintains a global Claude memory snapshot of all ongoing projects AND a per-project STATUS.md file in each project folder that captures last-session state, next actions, blockers, and key locations — so context survives long absences. Operates in two modes. WRITE mode triggers when the user says "save this", "memorise this", "snapshot this", "checkpoint [project]", "update your memory", "store this session", "remember all of this", or "save progress" — also proactively at the natural close of any productive session. READ mode triggers when the user says "where did I leave off", "catch me up on [project]", "restore context", "what was I doing", "[project] status", or returns to a project after a long gap. Works in both Claude Code (writes and reads STATUS.md directly in the project folder) and Claude.ai (updates global memory and emits STATUS.md content as an artifact for the user to save). Do not wait to be asked if a session was clearly productive.
howardxie-dev/ml-agent-skills/skills/tabular-ml-lab Skill
2★ repoUse when the user wants to use an LLM coding agent such as Codex or Claude Code to run a reproducible machine learning workflow on tabular CSV data, including data profiling, leakage checks, binary classification baselines, evaluation metrics, threshold reports, model cards, and final reports.
hec-ovi/blockchain-skill Skill
2★Operate a non-custodial blockchain wallet on-chain (EVM and Bitcoin), no exchange or MetaMask. Create or import a wallet, balances, send native/ERC-20/BTC, swap tokens, compile/deploy/call Solidity. Trigger on wallet, crypto, ETH, BTC, ERC-20, token, send, transfer, swap, trade, Solidity, contract, on-chain, testnet, mainnet, sepolia.
xFurti/leogriel/.leogriel/skills/leogriel Skill
2★ repoManage Agent Skills with the leogriel CLI: search, info, add, install, outdated, update, sync, import, audit, plugins, behavioral tests, completion, and portable manifest/lock workflows across Claude, Cursor, Codex, Grok, Gemini, Pi, and OpenCode. Use when the user mentions leogriel, agent-skills.json, agent-skills.lock, syncing skills, importing from npx skills, or skill portability across machines.
maoningge/claude-codex-review/skills/claude-codex-review Skill
2★ repoGet an independent second-opinion review from Codex (the `codex` CLI) on code or implementation plans that Claude just produced. Use this whenever Claude has just finished writing or substantially editing code, or has just drafted an implementation/design plan, and an independent cross-check would add value — and especially whenever the user says things like "have codex review this", "cross-check with codex", "get a second opinion", "codex 审一下", "codex 过一遍", or "让 codex 看看". Routes Claude's output to `codex review` (for code diffs) and `codex exec` (for plans), then evaluates Codex's findings critically rather than accepting them blindly.
pendig/rutebayar-agent-contributor Skill
2★Operate and extend Rute Bayar, the Go CLI and webhook daemon for Indonesian payment gateways. Use when setting up sandbox or production Rute Bayar flows, helping AI Agents create invoices and verify payments through the rutebayar CLI, testing webhooks/forwarding/reconciliation, preparing releases, or contributing a new provider adapter such as DOKU, Flip Business, Duitku, Xendit, or Midtrans.
lixuvip/codex-agent-orchestration-skill/skills/agent-orchestration Skill
2★ repoCoordinate Codex work across the current task, internal subagents, user-owned tasks, worktrees, handoffs, formal review gates, or recurring automation. Use for delegation, parallel roles, another Codex task, cross-repo work, QA/review handoffs, scheduled follow-up, or task inspection and continuation. Chinese triggers include 子Agent/子进程/子对话, 另一个对话接手, 并行处理, 分给QA, 多仓库收口, 巡检, 持续推进, 定时自动继续, 创建/读取/转交任务.
90sRehem/agent-skills/spec-driven Skill
no license2★ repoStructured planning pipeline with 4 adaptive phases (Specify, Design, Tasks, Execute) and persistent knowledge accumulation for development teams. Auto-sizes depth by complexity from quick fixes to multi-component features. Commands: /explore (understand problem + map context), /plan (write spec → design → tasks), /apply <spec> (execute a spec), /spec <desc> (full pipeline). Also triggered by natural language: 'plan this feature', 'specify X', 'help me structure this', 'how should we implement X', 'break this into tasks', 'explore this', 'entende o problema', 'vamos especificar', 'preciso planejar', 'implementar X', 'cria as tasks', 'quebra em tarefas', 'como vamos fazer X', 'me ajuda a planejar', 'initialize project', 'map codebase', 'quick fix', 'pause work', 'resume work'. Covers requirement traceability, atomic task creation with verification criteria, and session memory via docs/ knowledge base. Do NOT use for architecture decomposition analysis (use architecture skills), technical design docs (use technical-design-doc-creator), or general code review.
Codex managed worktree threads
daxiong888/codex-managed-worktree-threads Skill
2★Explicit-only workflow for coordinating one Codex App background thread per task in Codex-managed worktrees. Use only when the user invokes $codex-managed-worktree-threads. Do not use for natural-language-only parallel work requests, ordinary subagent delegation, manual git worktrees, tmux workflows, Codex CLI worker orchestration, or vague requests to work in parallel.
lil-lon/.claude/skills/agent-worktree Skill
2★ repoSpawn an isolated Claude Code sub-task in a fresh git worktree, split into a new tmux pane next to the caller, then clean up when done. Use this whenever the user wants to delegate a small, self-contained side task without disturbing the current session — e.g. they say "do this in a worktree", "spawn a sub-agent", "start a side task", "run this in the background", or ask for parallel work like "fix this typo while you keep refactoring". Always use this skill for those requests instead of editing files directly when the user's intent is to fork off a parallel job.
ShiyangZheng/scholarly Skill
no license2★Academic paper and thesis writing assistant - Guides users through Q&A to progressively build all sections of a research paper (Introduction, Literature Review, Methods, Results, Discussion, Conclusion). Works with any AI agent framework (OpenClaw, Claude Code, Cursor, etc.). trigger_when: 用户要写学术论文、学位论文(本科/硕士/博士)、期刊文章、研究报告;或提到「论文大纲」「文献综述」「研究方法」「论文结构」「Abstract」「Introduction」「Method」「Results」「Discussion」等关键词。
michpro/kicad-tools Skill
2★Automates KiCad EDA workflows — schematic analysis, PCB review and repair, DRC/ERC validation, autorouting, placement optimization, BOM generation, and manufacturing export — all via the `kct` CLI with `--format json` output. Use this skill when the user mentions KiCad file extensions (.kicad_sch, .kicad_pcb, .kicad_pro, .kicad_sym, .kicad_mod), refers to KiCad by name, or describes KiCad-specific tasks such as DRC/ERC checks, Gerber export, BOM generation, autorouting, placement optimization, or manufacturer rule validation (JLCPCB, OSHPark, PCBWay). Synonyms like 'schematic file', 'PCB layout file', or 'board design' also qualify when paired with a KiCad context. This skill wraps the CLI — it does NOT use the kicad-tools MCP server.
wakqasahmed/ai-engineering-workflow-skills/skills/engineering/ai-agent-pr-metadata Skill
2★ repoAdd GitHub-visible AI agent and AI code-review metadata and exact agent labels without adding AI attribution to commits, and scope Alibaba/Open Code Review's review surface to control LLM token spend. Use when configuring PR templates, issue/PR labels, PR creation commands, PR update comments, or Alibaba Code Review/open code review workflows to disclose the implementing agent, review tool, LLM model, run URL, or to add/tune a `.opencodereview/rule.json`.
AllanWessels/Bratan/docs/build-skills/agent-model-selection Skill
no license2★ repoPass model explicitly when dispatching sub-agents. Parent-default inheritance bypasses cost considerations; reserve stronger models for tasks that genuinely need them.
JakeLin/effective-claude-code/.claude/skills/effective-claude-code Skill
no license2★ repoApply Effective Claude Code book principles to a project's Claude setup. Use when someone asks to improve their Claude Code setup, apply best practices, or configure Claude Code for a project. Also triggers on "apply item-N", "apply chapter-N", "apply hooks/memory/settings/subagents", or "how should I configure Claude Code".
KayZhongyi/mailpilot-agent Skill
2★Crash-safe mail merge for AI agents. Give it a recipient list (optionally grouped by a column) and templates; it previews for review, sends in batches, is idempotent/resumable with an append-only ledger, and reconciles bounces. Great for status notifications, customer updates, and pre-sales follow-ups.
taikt/llm-wiki/skills/llm-wiki Skill
2★ repoManage a personal LLM-maintained wiki: ingest source documents, answer questions, create/update concept pages, lint the wiki, and configure projects. Use when the user says 'ingest', 'add source', asks a question about wiki content, says 'lint', 'audit', 'configure wiki', or mentions a filename to add. Supports configurable root folder so you can work with any project's wiki regardless of the currently open workspace.
Solonnikov/agent-skills/skills/agent-skill-creator Skill
no license2★ repoAuthors new skills for the agent-skills repo with consistent format — tight SKILL.md, frontmatter, references for depth, and a pre-merge checklist. Use when adding a new skill to github.com/Solonnikov/agent-skills, standardizing an existing skill, or auditing a skill's structure.
2654400439/Paper2Seminar/skills/paper-ppt-orchestrator Skill
2★ repoBuild complete fixed-layout academic-paper presentations from a PDF with bundled PPTX and one-page HTML templates. Use for full paper reading, complexity-aware slide budgeting, subsection coverage mapping, figure extraction, TikZ or Matplotlib evidence visuals, deterministic PPTX assembly, and final QA. Trigger for requests to turn a paper PDF into a discussion-group deck, especially 12-16+ page systems, security, networking, or measurement papers.
Universal prompt optimizer skill
Jupiter363/universal-prompt-optimizer-skill Skill
2★将模糊、宽泛、不完整的自然语言请求,通过主动追问澄清缺失信息后,标准化为 LLM 偏好的结构化 Prompt——包含角色、任务、背景、目标、约束、步骤、输出格式。适用于代码开发、Bug排查、重构、测试、文档、计划、研究、写作等 AI Agent 任务场景。当用户要求优化/改写/整理/结构化/澄清/增强 Prompt,要求先优化再执行、先把需求说清楚,或输入"帮我修一下bug""帮我优化这个页面""帮我写个文档""帮我规划一下""帮我重构一下"等缺乏具体目标与边界的中文口语化指令时触发。已非常具体明确的任务指令不触发。
zacharygcook/agent-skills/skills/agent-readiness Skill
2★ repoAudit, score, compare, report, and iteratively improve how safely and effectively coding agents can work in a software repository. Use for read-only agent-readiness audits, first-class HTML/PDF readiness reports, readiness levels or percentages, Factory-compatible comparisons, AGENT_READINESS_PREFERENCES.md setup, selecting remediations, or autonomous one-criterion-at-a-time improvement loops.
The same file is in 1 other repository
MadewellRD/skills-lab/dist/skills/ai-engineering-command-desk/agent-architecture-desk Skill
2★ repodesign AI agent architecture including planning boundaries, execution loops, memory and state strategy, tool routing, approval gates, retries, delegation, and halt behavior.
unshDee/proofrag/skills/proofrag Skill
2★ repoEvaluate a RAG or LLM app. Use when the user wants to test, score, benchmark, or catch regressions in a retrieval/RAG/LLM system, generate an evaluation/golden dataset from their docs, measure hallucination/groundedness/correctness, or gate CI on answer quality. Generates a golden set from the user's own corpus, runs LLM-as-judge plus retrieval metrics, and produces a shareable HTML scorecard.
cueapi/cueapi-mcp/ai.cueapi/mcp MCP server
2★ repoSchedule agent work and report write-once outcomes via CueAPI from any MCP host.
jmagar/apprise-rmcp/ai.dinglebear/apprise-rmcp MCP server
2★ repoRust MCP server and CLI for Apprise notification fan-out across dozens of delivery backends.
jmagar/gotify-rmcp/ai.dinglebear/gotify-rmcp MCP server
2★ repoRust MCP server and CLI for Gotify push notifications and message management.
jmagar/arcane-rmcp/ai.dinglebear/arcane-rmcp MCP server
2★ repoRust MCP server and CLI for Arcane Docker and container management.
jmagar/ytdl-rmcp/ai.dinglebear/ytdl-rmcp MCP server
2★ repoRust MCP server and CLI for yt-dlp search, downloads, metadata, transfers, and Plex playlists.