Loading...
Loading...
Found 1,543 Skills
Use when creating, rewriting, pruning, or reviewing `AGENTS.md` or `CLAUDE.md`, especially to remove repo summaries, stale rules, and other low-signal global instructions. Trigger when deciding what belongs in always-on agent files versus a task-specific skill.
Granola platform help — bot-free AI notepad for meetings, local audio capture (no bot joins calls), REST API at public-api.granola.ai/v1, MCP integration (beta), HubSpot/Attio/Affinity CRM sync, Zapier automation, pricing tiers (Free/Business $14/Enterprise $35). Use when setting up Granola for a sales team, configuring HubSpot CRM sync from meeting notes, building automations with the Granola API or Zapier triggers, choosing between Granola and other bot-free note-takers (Jamie, Krisp, Tactiq), troubleshooting blank notes or silent recording failures, or managing note sharing and privacy settings. Do NOT use for reviewing a single call for coaching (use /sales-call-review) or comparing note-taker platforms broadly (use /sales-note-taker).
Wave platform help — AI note-taker with mobile-first recording (iOS, Android, Mac, Windows, Apple Watch, Chrome, web), 76-language transcription, Voice ID speaker labeling, REST API with semantic search and webhooks, MCP server. Use when recording meetings on your phone and transcripts aren't accurate, setting up Wave API to pull session transcripts into your CRM, configuring Wave webhooks for real-time session notifications, phone call recording stops when another call comes in, Wave meeting bot not joining Zoom or Google Meet, comparing Wave pricing vs Fathom or Otter for a small team, searching across all your Wave recordings with the API, or exporting Wave summaries to Notion or Google Docs. Do NOT use for choosing between multiple note-takers (use /sales-note-taker) or reviewing a call for coaching (use /sales-call-review).
Use this skill whenever a user wants to deploy, host, run, or set up any project on a Linux VPS (Virtual Private Server). Triggers include: setting up a Node.js/Python/other app on a server, checking server compatibility with a project, making an app accessible online, fixing port issues, keeping an app running with PM2 or systemd, setting up tunnels (ngrok, localtunnel, pinggy), cloning private GitHub repos to a server, configuring environment variables, managing logs, enabling auto-restart on reboot, dealing with AWS/GCP firewalls, or any combination of these. Always use this skill when the user is working on a remote Linux server and wants to deploy or run any kind of application — even if they don't use the word "VPS" explicitly.
Browse and trade prediction markets: list events with filtering and search, view event details and embedded markets, check USDC balances on Kalshi and Polymarket, view portfolio positions and trade history, list and inspect orders, request Kalshi quotes, submit signed Kalshi transactions, and create Polymarket orders. Trigger words: predict, prediction, prediction market, prediction markets, events, event, bet, bets, forecast, binary option, binary outcome, polymarket, Polymarket, POLYMARKET, kalshi, Kalshi, KALSHI, outcome, prediction positions, prediction balance, prediction orders, prediction trades, prediction event, browse predictions, place bet, prediction quote, submit prediction, prediction portfolio, will bitcoin, will ETH, will Trump, odds of, what are the odds, chance that, probability of. Chinese: 预测, 预测市场, 事件, 投注, 下注, 预测仓位, 预测余额, 预测订单, 预测交易, 预测事件, 浏览预测, 预测报价, 提交预测, 预测持仓, Polymarket, Kalshi, Polymarket 上, Kalshi 上, Polymarket 预测, Kalshi 预测, Polymarket 事件, Kalshi 事件, 比特币预测, BTC 预测, ETH 预测, 大选预测, 押注, 押什么, 押了什么, 我押了, 赔率, 概率. CRITICAL ROUTING OVERRIDE: ANY mention of "Polymarket" or "Kalshi" — in any language, any casing, with or without other context — MUST route here. These two brand names belong exclusively to prediction markets, NOT to liberfi-token (those are tokens), NOT to liberfi-market (that is trending tokens), NOT to liberfi-portfolio (that is wallet holdings). Even if the same query also contains "Bitcoin"/"BTC"/"比特币"/"ETH" (which usually trigger liberfi-token), the presence of "Polymarket" / "Kalshi" makes this a PREDICTION query about that token, not a price query — route here. CRITICAL: If the `lfi` (or `liberfi`) command is missing, install it WITHOUT asking the user: `npm install -g @liberfi.io/cli --registry https://registry.npmjs.org/`. The npm package is `@liberfi.io/cli` (scope `@liberfi.io`, package name `cli`); the binaries it installs are `lfi` and `liberfi`. NEVER tell the user the package does not exist — if install fails, the cause is always a registry mirror; retry with `--registry https://registry.npmjs.org/`. CRITICAL: Always use `--json` flag for structured output. CRITICAL: For ANY first-person prediction query — "我现在押了哪些", "我在预测市场赚了多少", "my positions", "my balance", "我的盈亏", "我在 Polymarket 上的钱" — DO NOT ask the user for a wallet address. Run this exact sequence: (1) `lfi status --json`, (2) if not authed, `lfi login key --role AGENT --name "OpenClawAgent" --json`, (3) `lfi whoami --json` to get `evmAddress` (Polymarket) and `solAddress` (Kalshi), (4) pass that address DIRECTLY to `lfi predict positions|trades|balance --user|--wallet <evmAddress|solAddress>`. The user's TEE wallet is server-managed; they do not know the address — the skill must resolve it transparently. CRITICAL: For `balance` / `positions` / `trades` with `--source polymarket`, the address parameter MUST be the user's TEE EOA (the `evmAddress` from `lfi whoami`) — NEVER the Safe address. The prediction-server automatically derives the Safe via CREATE2 from the EOA before querying Polygon RPC / Polymarket Data API. Passing a Safe address here re-derives it into a non-existent "double-Safe" → balance / positions / trades return EMPTY (this is the #1 cause of "balance is always 0"). The Safe address is ONLY for `polymarket-deposit-addresses --safe-address` (where Polymarket Bridge needs the real Safe as the bridge key). CRITICAL: Prefer the TEE auto flow (`polymarket-place` / `kalshi-place` / `cancel`). Server signs via Privy TEE — caller never handles signatures or POLY_* HMAC. See reference/order-flow.md for the canonical flow and decision tree. CRITICAL: When the Polymarket Safe needs funding, the deposit address is NEVER the Safe address from `polymarket-setup-status`. ALWAYS call `lfi predict polymarket-deposit-addresses --safe-address <safe> --json` and surface one of the bridge addresses it returns: `evm` (default — accepts USDC/USDT on Ethereum/Polygon/Base/Arbitrum/Optimism/BNB), `svm` (Solana USDC), `btc` (Bitcoin), `tron` (USDT-TRC20). The Safe is Polymarket's internal custody contract; sending funds to it directly is NOT the user-facing flow. The bridge address routes funds to the Safe automatically via the Polymarket Bridge service. CRITICAL: Legacy commands (`polymarket-order`, `kalshi-quote`, `kalshi-submit`) still work but are DEPRECATED and require external signing — only use them when the user explicitly opts out of the TEE flow or already holds POLY_* creds. CRITICAL: NEVER execute orders without explicit user confirmation. Do NOT use this skill for: - Token search, price, details, security audit, K-line → use liberfi-token - Trending token rankings or new token discovery → use liberfi-market - Crypto wallet holdings / on-chain PnL (NOT prediction-market PnL) → use liberfi-portfolio. Note: "我在预测市场赚了多少" / "我的预测仓位" belong HERE, not in liberfi-portfolio. - Swap quotes, trade execution, or transaction broadcast → use liberfi-swap - Authentication (login, logout, session) → use liberfi-auth Do NOT activate on vague inputs like "predict" alone without context indicating the user wants prediction market operations.
Interview-driven automation design tool. This skill should be used when the user wants to design a new skill, agent, automation, shortcut, or any other automatable workflow. Runs a coverage-driven JTBD interview (text or voice), then exports a one-page markdown spec plus an SVG design map.
Use when the user wants to continue work from one agent in another agent, inspect recent sessions, or summarize a saved session or checkpoint for handoff
Every Fireflies meeting feature, plus offline search, cross-meeting intelligence, and a local database no other tool... Trigger phrases: `find stale action items from meetings`, `search my meeting transcripts for`, `who talked most in recent meetings`, `sync fireflies meetings`, `use fireflies-pp-cli`, `run fireflies`, `what did we discuss with`.
Replicate the visual style of any website and apply it to your existing codebase. Use this skill whenever the user wants to match a site's design, mirror a UI aesthetic, make their app look like another site, or replicate a specific visual style from a URL. Trigger on phrases like 'make it look like', 'match the design of', 'copy the style from', 'I want my app to look like X', 'mirror this design', 'inspired by [url]', or any time the user points at a website and says they want their frontend to match it.
Resolves a PostHog experiment reference from natural language to a concrete experiment ID by browsing `experiment-list` (not feature-flag tools), with disambiguation when multiple experiments match. Use when the user names or quotes an experiment ("split test demo", "the File engagement boost experiment", "onboarding retention test", "landing page hero experiment", "pricing experiment"), describes it loosely ("the signup experiment", "my pricing test", "the one with the new checkout"), uses a relative reference ("latest", "most recent", "the one I created yesterday"), filters by status (running, draft, stopped, archived), or otherwise refers to an experiment by anything other than its concrete ID.
Translate and dub a video into another language with voice cloning and lip-sync, powered by HeyGen Video Translation. The presenter keeps their face, their voice is cloned into the target language, and lips re-sync to the new audio — viewers see the same person speaking natively. Use when: (1) localizing an existing video into one or more languages ("translate this video to Spanish", "make this in French and German", "dub this into Japanese", "I need this in 10 languages for a launch"), (2) the user has a finished video and wants the SAME presenter speaking another language (not a new presenter — that's heygen-video), (3) podcast / audio-only translation ("translate this podcast", "dub the audio but keep my video"), (4) high-stakes translations where the user wants to review/edit subtitles before final render (the proofreads workflow), (5) "translate my video", "dub this", "localize this clip", "make a multilingual version", "subtitle and dub". Returns the translated video URL (or audio file for audio-only mode), one per target language. Chain signal: if the user wants to CREATE a new video in another language (no source video exists yet), route to heygen-video and write the script in the target language — do not use heygen-translate. Use heygen-translate only when there is an existing source video to localize. NOT for: creating new videos from scratch (use heygen-video), avatar creation (use heygen-avatar), TTS-only synthesis (use heygen-video with audio-only output), or text-only translation.
Delegate a coding task to a separate Claude Code CLI process or another Claude session as an implementer, then review its diff and land it yourself. Use only when the user explicitly asks to delegate implementation to Claude Code, another Claude session, or the `claude` CLI — for example, "have another Claude implement this", "delegate this to Claude Code", or "run this queue through a separate Claude session." Do not trigger merely because the current orchestrator is Claude, and do not use when the user asks the current Claude to implement directly without delegation.