The catalog
609,344 rows, by stacks then stars
ProdaZhang/aigd-zh/aigd-concept Skill
1★ repoAIGD 阶段1 · 头脑风暴与顶层设计。当要给新游戏立意、定核心循环/平台/目标用户,或把游戏拆成系统清单+依赖图、起项目脊柱时使用。产出/修订 项目档案 + manifest(系统清单)。是 aigd 包的一员,方法论见 ../aigd/references/。
mtthsnc/tempest/plugins/tempest/skills/bcp Skill
1★ repoUse when brand or marketing work needs to be on-brand and traceable — set up a Brand Context Protocol (BCP) for a business, capture brand truth, or produce/score a deliverable (landing page, deck, email, ad copy) against the brand. Triggers on "set up a brand", "make this on-brand", "scaffold a BCP", "brand context protocol", "produce on-brand <thing>", "is this on-brand". Bridges the dev stack to the BCP template at github.com/mtthsnc/bcp.
comment-hq/skills/skills/code-review Skill
1★ repoRun one evidence-focused, risk-scaled official code review on a pull request and post the result. Only invoke when explicitly requested as `$code-review` or `/code-review`, or when `ship` selects it as the sole official-review fallback.
hztopcu/academic-research-skills-universal/ars_universal/assets/academic-paper-reviewer Skill
no license1★ repoMulti-perspective academic paper review with dynamic reviewer personas. Simulates 5 independent reviewers (EIC + 3 peer reviewers + Devil's Advocate) with field-specific expertise. Supports full review, re-review (verification), quick assessment, methodology focus, Socratic guided, and calibration modes. Triggers on: review paper, peer review, manuscript review, referee report, review my paper, critique paper, simulate review, editorial review, calibrate reviewer, reviewer calibration, measure reviewer accuracy.
kirklin/skills-latex Skill
1★Create polished, compile-tested LaTeX documents from curated kirklin templates and build them to PDF with per-page PNG previews. Use when the user wants to write, typeset, or compile a LaTeX document — especially an academic or research paper in NeurIPS/arXiv style (the neurips-paper template) — or says "make a LaTeX paper", "write a conference paper", "typeset this as a paper", "compile this .tex", or "turn this into a PDF paper". This is a template framework: pick a template under templates/, copy and customize it, then build with scripts/compile.sh (auto engine detection, latexmk, previews). New document types are added as templates.
Vertiso/memory-codex/vertiso-memory/skills/checkpoint Skill
1★ repoSave work-in-progress to Vertiso Memory without ending the session, preserving decisions and their reasons verbatim across every topic. Use when the user asks to checkpoint, save progress, capture this, or not forget something, or affirmatively accepts an agent's checkpoint offer. An agent may offer on a cadence or after a salient moment but must not perform the capture without the user's request or approval.
The same file is in 2 other repositories
Expert-Vision-Software/aurelia-expert/.agents/skills/notebooklm Skill
1★ repoComplete API for Google NotebookLM - full programmatic access including features not in the web UI. Create notebooks, add sources, generate all artifact types, download in multiple formats. Activates on explicit /notebooklm or intent like "create a podcast about X"
Elisedai1013/build-reliable-agent-skills-skill/skills/build-reliable-agent-skills Skill
no license1★ repoTurn a recurring Agent workflow, an existing Skill, or concrete failure records into a narrowly scoped and verifiable Skill charter, then diagnose failures into the correct layer: trigger, knowledge, tool, verification, or guardrail. Use when deciding whether a workflow deserves a Skill, defining what 'done' means, fixing under-triggering, converting repeated reconstruction into scripts, adding evidence-based gotchas, preventing false completion, adding safety controls, or evolving a Skill from real usage. Do not invent organization-specific facts or claim success without observable evidence.
liufeng1976/bossai-radar-lite/skills/bossai-radar Skill
no license1★ repoUse BossAI Radar Lite to inspect overseas business opportunities, verify public evidence, read commercial leads, prepare daily follow-up queues, and draft human-reviewed customer messages.
brian8a1/senso-agent-skills/skills/backlog-brief Skill
no license1★ repoBrief a product manager, designer, or engineer who is beginning work on one Senso Backlog item. Use when the user names a SEN reference, Opportunity, Solution, issue, branch, design task, or current work item and needs the relevant problem, Signals, Decision, Living Document, constraints, contradictions, and success expectation without an invented implementation workflow.
keithhegit/sci-using-kit/skills/literature-review-search Skill
1★ repoUse when finding references, building literature review tables, identifying up-to-date or highly cited papers, comparing research methods, verifying citation quality, mapping references to claims, or converting AI literature notes into SCI-ready bibliography inputs
frostime/slim-bad-skill/skills/slim-bad-skill Skill
1★ repoUse this skill when the user explicitly considers an existing SKILL.md bloated, mechanical, list-heavy, workflow-driven, or over-prescriptive and wants it substantially simplified without losing real capability. Recover the skill's core contract, validate the supporting contracts required by its actual use case, and remove accidental process, redundant derivations, false closed sets, and author preference presented as necessity. This is a repair skill, not a general SKILL-writing standard or blanket terseness pass.
ytfh44/gpu-optimize-skills/skills/gpu-code-optimizer Skill
no license1★ repoLoad this skill and follow it when planning, reviewing, or carrying out performance optimization for GPU compute, resource, or runtime-state workloads, especially when deciding which specialized optimization skills to load.
jdiazromeral/tars/skills/ask Skill
1★ repoTARS Answers from Raw Sources — a local-first second brain. Raw immutable captures, FTS5 search, and agent skills that answer with citations.
Fatih0234/pi-caveman/skills/caveman-commit Skill
1★ repoUltra-compressed commit message generator. Cuts noise from commit messages while preserving intent and reasoning. Conventional Commits format. Subject ≤50 chars, body only when "why" isn't obvious. Use when user says "write a commit", "commit message", "generate commit", "/commit", or invokes /caveman-commit. Auto-triggers when staging changes.
The same file is in 143 other repositories
eligapris/viral-clip-editor-skill Skill
no license1★Turn any long-form video into viral-ready short clips (TikTok/Reels/Shorts) or a single stitched recap. Use this skill whenever the user wants to: cut a video into clips, create viral shorts from a long video, extract key moments from a recording, add captions to a video, crop a video for mobile/vertical, make a recap or highlight reel from a podcast/interview/tutorial, download a YouTube video and edit it, or convert any horizontal video into vertical social-media-ready content. Also trigger when the user mentions: clip maker, short form content, reel maker, highlight reel, video recap, podcast clips, interview highlights, key moments extraction, or auto-edit video.
Vertiso/memory-cursor/vertiso-memory/skills/checkpoint Skill
1★ repoSave work-in-progress to Vertiso Memory without ending the session, preserving decisions and their reasons verbatim across every topic. Use when the user asks to checkpoint, save progress, capture this, or not forget something, or affirmatively accepts an agent's checkpoint offer. An agent may offer on a cadence or after a salient moment but must not perform the capture without the user's request or approval.
The same file is in 2 other repositories
stevederico/hon-maguro/skills/hon-maguro Skill
1★ repoEvaluate any project against fixed "Hon Maguro" quality bars via bin/maguro (init/links/eval), optional Lighthouse median for SCORE1, and a capped fix loop (max 2 rounds). Binary pass/fail — no partial credit. Use when the user says "hon maguro", "eval this project", "quality bar check", "is this project ready", "audit code quality", or wants a pass/fail verdict against Hon Maguro standards. Do not use for generic Lighthouse/PageSpeed-only or open-ended "inspect and fix" requests unless they ask for Hon Maguro / these bars.
huyu605/capability-radar/skills/capability-radar Skill
1★ repoDiscover, verify, score, and recommend current open-source AI and software capabilities across agents, agent skills, MCP servers, tools and APIs, software libraries, and deployable products. Use when a user asks what already exists for a capability goal, wants open-source alternatives, needs technology or build-vs-buy research, or wants a market scan before building or adopting software. Clarify requirements before searching, use live sources, cite primary evidence, and require explicit confirmation before installing or running candidates.
LikC1606/lab-meeting-report-skill/lab-meeting-report Skill
1★ repoCreate or update evidence-grounded Markdown documents for lab meetings from pasted notes, local research files, papers, experiment results, directories, or explicitly scoped Lark/Feishu documents, chats, Minutes, and meeting artifacts. Use for lab meeting reports, research progress reports, journal club notes, experiment retrospectives, mixed progress-and-literature reports, or publishing and synchronizing a verified Markdown report to a Lark cloud document rather than creating slides.
memoryplugin/agent-skills/skills/memoryplugin Skill
1★ repoGives the user one long-term memory shared by every AI tool they use, powered by MemoryPlugin. Use when a task or conversation starts and knowing the user's preferences, projects, people, or decisions would help; when the user references past context ("as I mentioned", "my usual setup", "you know me"); when durable new facts, preferences, or decisions emerge that future sessions will need; when the user asks to remember, recall, save, or forget something; or when they mention a past conversation in this or another AI (ChatGPT, Claude, Gemini, Grok). Requires the MemoryPlugin MCP server.
renchris/pyramid-principle-full/skills/pyramid-principle-full Skill
no license1★ repoApply Barbara Minto's COMPLETE Pyramid Principle methodology (2010 ed., all 12 chapters + appendices) to an input body — a draft, notes, analysis, data dump, problem statement, or presentation — through sequential deep-thinking working sessions with hard stops between processes, before AND after producing the pyramid-structured output. Use when asked to "apply the full pyramid principle", "minto this properly", "structure this document/report/presentation rigorously", to define+analyze a problem the McKinsey way, or whenever a high-stakes communication must survive the strictest logic audit. NOT for quick formatting passes (use the compressed pyramid-principle skill for that).
ValenFelizia/csdd Skill
1★Apply Collaborative Spec-Driven Development as a lightweight coordination and durable project-state protocol. Use when a repository contains canonical `.csdd/` project-state documents, making it CSDD-aware, or when the user or project instructions explicitly require CSDD.
fmanimashaun/claude-skills/skills/fidara-design Skill
1★ repoThe Fidara design system — how to build consistent, modern, responsive UI in Rails 8 + Hotwire + Tailwind CSS v4, brand-parameterised via brand packs. Load this WHENEVER building or reviewing UI: components (buttons, cards, forms, nav, modals, tables, badges, alerts), page layouts, tokens/theming, dark mode, responsiveness, or brand/logo usage. It defines the token architecture (brand primitives → semantic roles → fluid scale), the layout primitives you compose instead of ad-hoc flex/grid, the component catalog with variant/size/state, the Stimulus interaction patterns, and the responsive doctrine. Consistency is enforced here, not left to taste — and **art direction** is taught rather than left undefined: visual hierarchy, per-surface aesthetic intent, and the difference between a surface that is correct and one that is considered.
duandigi/duandigi-growth-marketing-skill/skills/account-connection-planner Skill
1★ repoUse this skill when planning how users, organizations, and projects should connect external marketing, analytics, advertising, social, CRM, commerce, or website accounts without collecting provider passwords.
promptclickrun/purpose-check Skill
1★Use when ambiguity about the audience, intended outcome, constraints, consequences, or materially different approaches could change the deliverable; when the requested output likely does not fit the underlying goal; at consequential mid-task forks; or when the work touches consequential domains such as money, privacy, safety, authorization, legal constraints, or irreversible actions.
Conradgui/skill-creator-pro/skills/skill-creator-pro Skill
1★ repoDesign, build, validate, and publish production-ready Codex skills. Use when Codex needs to turn a repeated workflow or new capability into a new skill, organize its scripts/references/assets, forward-test it before first release, or package it as an installable repository.
fxw-labs/x-to-wechat Skill
no license1★通过 opencli 抓取 X(Twitter) 最新热门新闻/trending 话题,自动生成爆款微信公众号文章(不少于1500字),并通过 browser-use 浏览器自动化发布到公众号草稿箱。触发词:热门新闻转公众号、X转公众号、twitter转公众号、爆款文章生成、热点文章发布、公众号自动发文、trending转文章、opencli抓热点。
eligapris/story-cut-editor-skill Skill
no license1★Turn a long-form video (yt-dlp URL or local upload) into one tightly-edited highlight/summary video: find the moments that matter, cut them, caption them, crop for mobile virality, and join into one story with transitions. Use when the user wants a long video turned into a highlight reel, a "best moments" cut, a summarized version, or key clips joined into one video — for isolated standalone clips alone, video-analyzer + ai-viral-scorer + social-clip-captioner is enough. Triggers on "summarize this video", "make a highlight reel", "cut the best parts and join them", "turn this into one recap video", "pull the key moments and combine them". Read before hand-building a download+transcribe+cut+caption+join pipeline from scratch — encodes the working stack (yt-dlp with bot-detection handling, silence-chunked Whisper, transcribe-once-slice-many captioning, face-priority crop with blur-pad fallback, join step that verifies clips before concatenating) from real failures.
diodames/orphan-words Skill
no license1★Prevents orphan words and awkward last-line wraps using CSS text-wrap (balance/pretty/stable), editorial , and nowrap phrase binders. Use when editing prose or headings, reviewing typography/line breaks, or when the user mentions orphans, widows, last-line wrapping, , non-breaking space, text-wrap, text-wrap-style, balance, pretty, or stable.
andthezhang/lark-simple-skill/skills/lark Skill
1★ repoLark and Feishu operations using lark-cli. Use for Lark, Feishu, or lark-cli tasks.
markthebault/code-wiki/skills/build-code-wiki Skill
1★ repoGenerate, refresh, and validate agent-oriented repository documentation in .code-wiki/wiki.json and managed source-file headers. Use for initial Code Wiki generation, incremental documentation refreshes, stale-documentation checks, or Code Wiki validation in a Git repository.
k97/skills/skills/apple-appicon Skill
1★ repoGenerate Apple platform app icons (iOS, iPadOS, macOS, visionOS) from one source image, with Apple's Human Interface Guidelines as the north star and first-class Tauri support. Validates the source first — square aspect, >=1024px, alpha, sRGB/P3, 8-bit — then generates AppIcon.appiconset (with dark/tinted variants), .icns via iconutil with Apple's rounded-rect margins for legacy macOS, visionOS layered image stacks, and Tauri's full icon set including the dock-icon margin fix. Dependency-free on macOS (sips, iconutil, Swift/CoreGraphics). Use for "generate app icons", "app icon from this image", "make an .icns", "AppIcon.appiconset", "tauri icon", "dock icon looks too big/square", "App Store icon". Roadmap: Flutter, React Native, Ionic.
ZeusSheen/ppro-agent-bridge/.claude/skills/ppro-bridge Skill
no license1★ repoDrive Adobe Premiere Pro through the ppro-agent-bridge CLI for transcript-driven, closed-loop video editing. Use when asked to inspect a Premiere project, transcribe footage (local Whisper), read transcripts, export frames or OTIO timelines, import media or transcripts, place clips or sentence-level cuts on the timeline, or assemble a rough cut / best-take edit from talking-head footage. Requires the broker running and the Agent Bridge UXP panel loaded in Premiere.
HomoLand/xbloom-studio-brew/skills/xbloom-studio-brew Skill
1★ repoDesign bean-specific hot pour-over and manual-over-ice Americano-style flash-brew recipes for xBloom Studio, research cited roaster/cafe/xPod references, import/query/sync a private app-visible coffee/tea catalog, preview or explicitly add/delete local account recipes, import App brew history, keep a local brew journal for dial-in, run guarded Omni Tea Brewer recipes, dial in by taste, and operate bundled local BLE for diagnostics, settings, scale, grinder, temperature/volume water, recipe load, A/B/C presets, monitoring, cancel, persistent pause/resume, and explicitly gated physical starts. Use for xBloom Studio, Omni Dripper, xPod/NFC Recipe Cards, Omni Tea Brewer, official or saved coffee/tea recipes, iced coffee, C40 conversion, WAIT troubleshooting, electronic-scale readings, standalone grinding/water, brew history, or direct xBloom Bluetooth control.
cosmefae/claude-skills/skills/ai-daily-brief Skill
1★ repoGenerates a daily brief of the top AI-world news, deduped across a curated source list, with structured themed output.
Vertiso/memory-claude/vertiso-memory/skills/checkpoint Skill
1★ repoThe official Vertiso Memory plugin for Claude Code.
The same file is in 4 other repositories
SudewaJay/apple-design-skill/apple-design Skill
1★ repoApply Apple's current design language (Human Interface Guidelines + Liquid Glass, mid-2026) to any UI work — app interfaces in the iOS/iPadOS/macOS style, and marketing/landing pages in the Apple.com style (huge headlines, alternating light/dark sections, scroll-driven product reveals). Use this skill whenever the user asks for "Apple style", "Apple-like", "HIG", "iOS style", "macOS style", "Liquid Glass", "iOS 26 style", "Cupertino", a "clean premium minimal" landing page, or a product page that feels like an iPhone/Mac launch page — even if they don't say "Apple" explicitly but describe frosted glass, floating tab bars, pill buttons, giant centered headlines, or scroll-scrubbed product animations. Also use when reviewing/critiquing a design for Apple-platform fit. Works for HTML/CSS/JS prototypes, React + Tailwind, and Figma design guidance.
cskwork/superqa-skill Skill
1★Browser QA for any website. Use when the user says QA or browser test; gives a URL to verify; names a known domain or feature to re-QA; wants a regression sweep after a feature lands; asks for a quick smoke check; wants to record a test by clicking, schedule one, or open the QA dashboard; or needs QA against a local stack because the shared environment is down or the cases are destructive.
transon-org/transon-authoring-skill/skills/transon-authoring Skill
1★ repoAuthor engine-valid Transon JSON transformation templates for a described data transform. Use when the user wants to create, generate, write, or fix a Transon template (a Transon transform), or to reshape, flatten, map, project, or restructure JSON data into another JSON shape via the Transon engine — especially when input/output sample pairs are available to verify against.
ambiguouStr/anki-review Skill
1★Quiz the user on their due Anki flashcards (synced from AnkiWeb) directly in chat, using Anki's real spaced-repetition scheduler that syncs grades back to AnkiWeb. Works with any deck or subject. Use when the user says "quiz me", "anki", "review my cards", "flashcards", "背单词", "复习卡片", or asks to memorize/practice cards.
flowleap-ai/flowleap-plugins/plugins/flowleap/skills/flowleap-academic Skill
1★ repoSearch academic literature (Semantic Scholar + arXiv) through the FlowLeap backend with per-source and publication-year filters. Trigger when an agent needs scholarly papers as prior art or technical background — topic searches, restricting to arXiv or Semantic Scholar, or bounding results by publication year. For OpenAlex-backed NPL search see flowleap-npl.
The same file is in 1 other repository
Paldom/llm-council-skills/.claude/skills/add-skill Skill
1★ repoAuthors or improves an Agent Skill in this repo via the eval-first workflow - scope one purpose, write trigger evals, draft SKILL.md, validate, iterate. Use when the user asks to create, add, write, refine, or fix a skill ("add a skill for X", "improve triggering of Y"). Not for scaffolding a new repository or editing infrastructure (hooks, CI, the validator/scripts, docs).
The same file is in 11 other repositories
voacatofe/skill-db-paid-media Skill
1★Global dlt-first AI skill for Meta Ads and Google Ads PostgreSQL warehouses
dralgroup/mlatom-skills/skill-template Skill
1★ repoOne sentence on what this skill helps an agent do.
linxumoney/earnings-signal-reviewer/skills/earnings-signal-reviewer Skill
1★ repo分析公司季度或年度财报,区分已披露事实、管理层叙事和分析推断,检查关键指标预期差、增长质量、利润与现金流、一次性项目、指引变化、风险和后续验证点。适用于财报分析、业绩电话会、季度更新、盈利质量、经营变化和投研简报;不提供买卖建议。
ceilidhboy/skills/skills/action-pattern Skill
1★ repoCreating and using action classes to encapsulate business logic. Actions replace traditional Service classes with reusable, composable, single-responsibility classes that follow a consistent `execute()` convention.
SwiftyJunnos/design-ui-prototypes/skills/design-ui-prototypes Skill
no license1★ repoUse when a user asks to design or redesign an application screen, compare visual UI directions, turn an approved direction into an interactive prototype, resume from an existing UI artifact, or prepare approved UI work for implementation.
dekrezz/ultraloop/skills/ultraloop Skill
1★ repoCompletion discipline for coding agents: a task is done only when the result is verified and demonstrably works — not when the code is written. Overrides the default "code written = done" and right-sizes its process to the task. For a small, self-contained change it delivers the solution immediately on sensible defaults with its assumptions stated, without unnecessary questions or process overhead. For a feature or multi-step effort in a real project it inspects the repository first to answer its own questions, turns whatever the code cannot reveal into explicit flagged assumptions rather than interrupting, sets an explicit Definition of Done and an iteration budget (max turns) itself, then executes autonomously — looping change → tests → build → lint → live check (Playwright / Computer Use) → fix until every criterion passes — before assembling a clean pull request that passes CI/CD. Use this skill whenever the user invokes "ultraloop" or "/ultraloop", or asks to carry a task through to a verified, working result on its own — for example "make it and verify it works", "don't stop until it works", "get CI green", "work on it while I'm away", or "verify it actually works". It is not needed for a pure factual question or a trivial one-liner that can be answered inline.
hunterbohm/skills/skills/business/ask-hormozi Skill
1★ repoApply Alex Hormozi's published business frameworks with constraint-first diagnosis and source-backed attribution. Use when the user asks what a Hormozi term or framework means; what Hormozi would do in a business situation; wants an offer, funnel, script, numbers, or draft analyzed or produced through his frameworks; or wants a Hormozi attribution or current position verified against public sources.
LcpMarvel/macos-app-icon-designer/skills/macos-app-icon-designer Skill
1★ repoDesign, critique, refine, validate, integrate, and package production-ready macOS application and menu bar icons. Use for macOS App Icon concepts, 1024px masters, layered Icon Composer artwork, Liquid Glass variants, Xcode AppIcon catalogs, Tauri or Electron desktop apps, legacy .iconset/.icns export, Finder/Dock previews, status item or tray template images, small-size legibility checks, or reviews against Apple guidance.
linkc-skills/clarify-with-questions Skill
1★通过一次一个自适应问题,帮助用户澄清模糊想法、发现关键假设、明确目标与取舍,并把零散信息转化为可执行、可表达或可复用的成果。适用于用户明确调用“clarify-with-questions”,或提出“采访我”“一个一个问”“先问清楚再写”“通过提问帮我梳理”“帮我把这个想法想清楚”等需求,以及产品、内容、策略、研究和个人复盘场景。用户显式指定 `--voice`、要求适合语音交流,或运行环境明确告知当前为语音会话时,使用语音友好的提问方式。
pathak-prashant/cash-framework/skills/cash-framework Skill
1★ repoUse this skill to evaluate architectural decisions and route each component to its optimal execution environment — traditional Code, GenAI, or Hybrid — using the C.A.S.H. framework (Constraints, Ambiguity, Sequencing, Heterogeneity).
yoelgal/better-dev/skills/autonomous-loop Skill
1★ repoPortable dev practices, packaged as skills - that run inside the coding agent you already use (Claude Code, Codex, pi, hermes). Not an agent, framework, or provider layer.
Rmalnoult/craft-skill Skill
1★Generate distinctive, production-grade frontend code that looks authentically designed, not AI-generated. Framework-agnostic. Use when the user asks to build, design, or style a web component, page, landing page, dashboard, or any frontend UI ("craft a page", "make this look good").
freddyjpp/supernoteOCR-skill Skill
1★Use when the user asks to sync, import, transcribe, or process Supernote .note files into their markdown notes (an Obsidian vault or any markdown folder). Triggers include phrases like "sync supernote", "pull my supernote notes", "process the supernote folder", or any request to convert handwritten Supernote notes into markdown.
mehul0810/loop-creator-skill/loop-creator Skill
1★ repoInterview users and inspect their setup and business to design, recommend, implement, pilot, and improve version-aware, token-efficient AI work loops with roughly 90–95% autonomous execution and explicit human approval for critical or destructive actions. Use for a solopreneur, one product, multiple-product portfolio, team, or organization; for an agent loop, automation workflow, recurring process, Ralph-style coding loop, evaluator/refiner workflow, AI operating system, LOOP.md/RALPH.md control document; or when adapting cadence, model choice, reasoning depth, permissions, and budgets to a particular business, AI tool version, workflow, risk profile, and working style.
he8um/airtable-skills/airtable-skills Skill
1★ repoprofessional airtable architecture and implementation guidance for designing, auditing, refactoring, documenting, and automating airtable bases. use when a user asks to build or improve airtable schemas, tables, fields, linked records, formulas, views, interfaces, forms, permissions, automations, scripts, api integrations, sync workflows, migration plans, quality checks, or agent-ready implementation prompts.
thisco/sdd-lifecycle Skill
1★Orquestra o ciclo completo de Spec-Driven Development (SDD) com rigor proporcional ao risco: classifica cada mudança em tiers (0 trivial / 1 pequeno / 2 estrutural) e conduz spec → plano → implementação TDD → revisão adversarial → destilação de memória. Reconhece a governança do repositório (constituição AGENTS.md, steering, memória, Arquitetura Viva) quando presente e degrada graciosamente quando não. Autocontida e agnóstica de agente de IA: funciona com qualquer agente baseado em LLM.
Justin-Ju-0413/aihome/data/sample-agents/doc-writer Skill
1★ repoGenerates comprehensive documentation for code projects, including README files, API docs, and user guides.