Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Productivity
Auto-grow a pro-workflow wiki by running a budget-capped BFS research loop over pluggable source fetchers (web, arXiv, GitHub). Each iteration pops a seed from the queue, fetches sources, drafts a wiki page, dedupes claims against existing pages, enqueues follow-up seeds. Halts on budget cap, depth cap, or convergence. Use when the user says "research <topic>", "grow the <slug> wiki", "auto-research", or wants a knowledge base that builds itself overnight.
Extract YouTube transcripts and produce structured concept analysis with multi-level summaries, key concepts, takeaways. Uses youtube-transcript-api with yt-dlp fallback. Triggers on: "analyze youtube video", "youtube transcript", "summarize this video", "extract concepts from video", "video key points", or any youtube.com/youtu.be URL
Use when a task requires interacting with a website beyond just reading it — clicking elements, filling forms, submitting data, navigating through multi-step flows, taking screenshots, or any workflow where the user needs a real browser with actions like click, type, scroll, or select. Also use for managing browser bookmarks, history, or tabs. Trigger whenever the user mentions browseros, browseros-cli, or BrowserOS. Do NOT use when simply fetching or reading page content would suffice — use curl, fetch, or WebFetch for that instead.
Challenges plans, designs, and decisions from multiple adversarial perspectives to surface blind spots, hidden assumptions, and vulnerabilities before they cause real damage. Use when testing plans or decisions for blind spots, need adversarial review before launch, validating strategy against worst-case scenarios, building consensus through structured debate, identifying attack vectors or vulnerabilities, or when groupthink or confirmation bias may be hiding risks.
Analyzes your recent Claude Code chat history to identify coding patterns, development gaps, and areas for improvement, curates relevant learning resources from HackerNews, and automatically sends a personalized growth report to your Slack DMs
Sắp xếp gọn kho tài liệu LOCAL (llmwiki/html/ + wiki/sources/draft/) khi phình to. Ba tầng — PROMOTE bản chất quý lên wiki ADR/concept (travel, sống sót clone), ARCHIVE render ephemeral (cặp proposal đã xong, bản bị thay thế, report cũ → archive/), KEEP canonical — rồi RE-INDEX đầy đủ. Trigger khi "dọn docs", "sắp xếp/merge tài liệu", "tidy docs", "html/draft quá nhiều", "/docs-curate".
Calculates probability-weighted averages of all possible outcomes to enable rational decisions under uncertainty. Covers scenario identification, probability estimation, payoff quantification, and risk-adjusted interpretation. Use when comparing risky options (investments, product bets, strategic choices), prioritizing projects by expected return, assessing whether to take a gamble, or when user mentions expected value, EV calculation, risk-adjusted return, probability-weighted outcomes, or decision tree.
Content-adaptive wisdom extraction that reads content first, detects which wisdom domains are present, and builds custom sections around them, with five depth levels and mandatory contrarian takes; pulls YouTube via fabric and articles via WebFetch. USE WHEN extract wisdom, analyze video, analyze podcast, extract insights, key takeaways, summarize interview, distill content. NOT FOR static Fabric extract_wisdom pattern (use Fabric).
Covers the full meeting lifecycle for engineering managers — produces guidance on whether to schedule a meeting, how to run it well, how to protect team focus time, how to kill recurring waste, and how to evaluate a past meeting from a transcript or description. Use when the user says "too many meetings," "meetings are a waste of time," "how do I run this meeting," "meeting agenda," "meeting culture," "nobody comes prepared," "meetings go nowhere," "how do I decline meetings," "distractions," "focus time," "engineers can't focus," "context switching," "protect engineering time," "review this meeting," or "transcript."
Helps engineering managers assess and improve team health across morale, cohesion, delivery culture, and engagement — produces Google's 5 Factors (Project Aristotle), a 4-state team health diagnosis (Falling Behind / Treading Water / Repaying Debt / Innovating), a 5-zone intensity model, the Engagement Stack, the Trust Battery, Teamicide patterns (Peopleware), a blameless postmortem format, and a library of team activities organized by driver. Use when the user says "team morale," "team is struggling," "burnout," "engagement," "attrition risk," "psychological safety," "team dynamics," "something feels off," "team culture," "team is unhappy," "retros aren't working," "team isn't working hard enough," "ideas for team activities," or "how do I run a team offsite." Do NOT use for individual performance concerns (use `managing-high-performers`), team staffing or hiring (use `team-composition`), or individual motivation interventions (use `engineer-motivation`).
Render a self-contained HTML viewer for a pro-workflow wiki. Pages, sources, claims, seed queue, page-link graph and full-text search all in one file. No external dependencies, no JS framework, S3-uploadable. Use when the user wants to browse a wiki visually, share its current state with someone, audit research progress, or hand off a knowledge base. Inspired by Thariq Shihipar's "Unreasonable Effectiveness of HTML" — favors information density and shareability over markdown-only outputs.
This skill should be used when users need help designing content workflows, creating process documentation, implementing automation rules, designing approval processes, or optimizing content pipelines. It activates when users ask about workflow design, process documentation, automation, approval workflows, or content pipeline optimization.
Automatically save and restore development context to minimize cognitive load when resuming work. Use when switching tasks, taking breaks, or returning after interruptions. Captures mental state, file locations, and next steps. Designed for ADHD developers with high context-switching costs
Conduct in-depth, multi-step research on a given topic by decomposing queries, finding diverse sources, cross-referencing findings, and synthesizing a comprehensive report. Use when the user requests deep research or provides relevant inputs for this workflow.
Structured multi-angle exploration running 2-8 sequential passes over the same problem, each through a different scientific lens, to surface hidden requirements and edge cases invisible from one angle; each pass yields new ISC criteria. USE WHEN iterative depth, explore deeper, multi-angle analysis, surface hidden requirements, blind spot check, what am I missing. NOT FOR scope/zoom analysis (use ApertureOscillation).
Generate a structured launch postmortem report covering goal attainment, reasons for deviations, lessons learned, and follow-up action items (including responsibility attribution and improvement owners). Use this skill when the user says "写复盘", "上线复盘", "项目复盘", "复盘报告", "postmortem", "版本回顾", "迭代总结", "上线总结", or when the user provides launch data / process records / an issue list and asks for a structured summary. Also applies to: incident postmortems (RCA reports after P0/P1 incidents), OKR retrospectives, quarterly retrospectives, A/B experiment retrospectives. Not applicable to: pure requirements writing (use pm-prd-writer), pure prioritization (use pm-prioritization-engine), weekly/daily reports (use status-report).
Run a structured, business-value-first product-planning interview that turns a fuzzy idea into a build-ready spec — BEFORE any code is written. Use this the moment someone is still figuring out WHAT to build and says things like "I have an idea for an app, help me think it through", "help me plan/scope/design my product", "where do I start?", "let's spec this out before I write code", or "turn this into a PRD". Covers any new app, site, SaaS, MVP, marketplace, or feature. Trigger on the planning intent alone — the user need NOT mention Strapi, a CMS, or any tech; a Strapi v5 + Strapi Cloud backend (overridable, with your choice of frontend and auth) is assumed by default. It nails value, users, and requirements before any tech decision, then produces six markdown files ending in a build spec you can hand to any coding agent (Claude Code, Cursor, etc.). Do NOT use once a spec or project exists and the user wants to scaffold, configure, migrate, add pages, wire auth, or debug — those are build tasks, not idea-stage planning.
WeChat Official Account article aggregator. Uses the mptext.top API to batch-fetch the latest article lists from specified accounts, download article content, and parse it into Markdown/HTML/plain text. Supports fetching by account name or fakeid and includes 8 popular AI technology accounts. Use cases: retrieving the latest articles from specified accounts, batch scraping content from multiple authors, aggregated reading of account articles, and collecting tech news. Keywords: official account article retrieval, WeChat article scraping, account aggregation, article collection, mptext API.
Unified entry point for Feishu cloud space, covering Drive enhanced upload/download and asynchronous import/export, basic file/media operations, wiki knowledge base nodes and space management, document comments, collaborators, and public permissions. Must be used when the user mentions Drive files, folders, large-file chunking or resumable transfers, asynchronous tasks, directory mirroring, knowledge base, wiki, assets, comments, sharing permissions, collaborators, public links, share passwords, ownership transfer, or permission requests. Wiki belongs to this Skill only when managing spaces, node structures, or members; it is explicitly forbidden for reading or summarizing Wiki/document bodies, which use feishu-cli-docs. Meeting recordings, Magic Share, and verbatim transcripts use feishu-cli-meetings; Drive-scope search belongs to this Skill; global document/message/app search uses feishu-cli-platform.
Use when the user wants to organize, tidy, or restructure a messy directory (Downloads, Desktop, documents folder). LLM-powered file organizer that scans content, proposes a structure, and executes moves in chunks with a mandatory user confirmation gate.
A set of resources to help me write all kinds of internal communications, using the formats that my company likes to use. Claude should use this skill whenever asked to write some sort of internal communications (status reports, leadership updates, 3P updates, company newsletters, FAQs, incident reports, project updates, etc.).
Launches a simulated meeting with several expert personas to analyze a topic from diverse perspectives, make a decision, and propose a solution for the user to approve before implementation. Does not write or commit code without explicit user approval; optionally, it can publish the analysis to a linked GitLab or GitHub issue. Use when the user wants to debate an approach, get multi-expert opinions, or receive a recommendation before implementation. For a fully autonomous meeting that implements and delivers directly (commit + MR/PR) without an approval step, use fast-meeting instead
Cross-platform real-time web research and search via an OpenAI-compatible Grok endpoint, returning JSON with content and sources. Use for version checks, API and docs verification, troubleshooting, and any time-sensitive facts on Windows, macOS, or Linux.
Search scholarly works across arXiv, conferences, journals, and books via OpenAlex. Returns citation counts, reconstructed abstracts, venue, open-access PDF links, and external IDs (arXiv/DOI/PMID). No API key required. Use when the user says "search scholar", "search papers", "find papers with citations", "papers by author", "cross-discipline search", or "papers from <venue>". Complements arxiv-search (preprints-only).
Search the web and scrape pages using the local tool stack: SearXNG (meta-search), Lightpanda (fast headless fetch), and Agent-Browser (full browser automation). This is your DEFAULT web skill — use it whenever you need to look something up, research a topic, fetch a webpage, extract content from a URL, check current information, find documentation, do competitive research, or answer any question that benefits from live web data. Triggers on any form of: search for, look up, google, find out, research, what's the latest on, fetch this page, scrape this site, check this URL, pull info from, web search, or any task where current web information would improve your answer. Even if the user doesn't explicitly ask you to search — if answering well requires current info you don't have, use this skill. NOT for interactive browser automation like form filling or clicking (use [[agent-browser]] or [[browser-use]]).
Answer citation questions about an Obsidian vault through ZotLit: which notes cite a work, what a document cites, and why a citation is broken. Use whenever a user asks who cites something, what a note references, where a work is discussed, or why a citation does not resolve.
Automatically collect and summarize daily AI industry news, trends, and hot topics from platforms like GitHub (trending repos), X/Twitter (AI influencers/hashtags), and AI news aggregators. Use this skill when the user asks for "today's AI news", "AI industry updates", "what's trending in AI", or wants a daily digest of AI developments.
Platform foundation capabilities for Feishu CLI only; not a general fallback for all Feishu requests. Covers config initialization, OAuth login and Token/Profile management, doctor diagnostics, OpenAPI schema lookup, generic api passthrough, global search, and user and department queries. Must be used when the user mentions logging in to Feishu, Device Flow, scope, User/Tenant Token, Token expiry, profile, doctor, 99991672/99991679, querying an API path/parameters/scope, raw api, calling an unwrapped OpenAPI, or searching docs/messages/apps or querying users, emails, or departments. Explicitly not for document bodies, Drive files, messages/group chats, Sheet/Bitable, whiteboards/boards, calendar/tasks/approval/attendance/OKR, email, or meetings/Magic Share; those business operations go to the corresponding domain Skill. "Global search" here refers only to `search docs/messages/apps` and does not include queries within business domains such as approval, meetings, or email. Business approval definitions/instances/todos use feishu-cli-work; meetings/Magic Share operations use feishu-cli-meetings. However, when explicitly querying a schema or calling an unwrapped raw OpenAPI, this Skill still applies even if the endpoint belongs to approval/vc.
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.