Browse reusable Agent Skills, each with a clear purpose and practical guidance.
Search people and companies, inspect contact intelligence, and review company signals in RocketReach — funding, growth, size, and tech stack. Use when users want to find decision-maker contact info, research company profiles, or build prospect lists.
Stack Exchange API integration with managed OAuth for Q&A knowledge automation. Search questions, retrieve answers, browse tags, manage user profiles, track reputation, list collectives, and analyze community activity across all Stack Exchange sites. Use this skill when users want to search Stack Overflow for solutions, find answers to technical questions, browse tags and badges, track user reputation, or analyze community activity patterns.
服务于 `/genesis` Step 3「技术选型」：以 ATAM 与 12 维加权矩阵评估候选栈，产出可追溯的对比结论与 ADR 升格素材；不写正式 ADR 文件（落盘在 Step 5）。优先读本 SKILL 旁 `references/`。
Scrape, crawl, or extract structured data from one or more URLs via the `webreaper` CLI. Outputs clean Markdown by default; JSON when a schema is given, or schema-free JSON when an LLM endpoint is configured (extract by natural-language prompt, no CSS selectors). Maps a site's URLs in one call. Handles JS-rendered pages and bot-protected sites (Cloudflare, DataDome, PerimeterX) via auto-escalating stealth. Use this skill whenever the user asks to: - scrape, crawl, or extract from a URL or site - get clean Markdown of a webpage (for further processing, not a summary) - pull specific fields from one or many pages (via a CSS schema, or by a natural-language prompt when an LLM endpoint is configured) - enumerate / discover URLs on a site - read a JS-rendered single-page app - scrape a site that's blocking direct requests Trigger phrases include: "scrape <site>", "crawl <site>", "extract <data> from <url>", "what's on <site>", "what pages does <site> have", "give me the markdown of <url>", "convert <url> to markdown", "pull <field> from <url>", "save <article> as markdown", "build a scraper for <site>", "read <url> into context", "this site is blocking me", "Cloudflare-protected site". Prefer this over the built-in WebFetch whenever the user wants: - Clean Markdown output to work with downstream (not just a summary in chat) - Structured field extraction via schema - Multi-page or site-wide work - JS-rendered or bot-protected sites WebFetch is the right tool only for "read this single URL and tell me about it" — when the output is a conversational answer, not data. Anything that produces an artifact (file, structured record, multi-URL result) belongs here.
Recursively builds a knowledge graph of mathematical concepts and ML algorithms. Use when the user asks to "build a knowledge graph", "decompose concepts", "map dependencies between algorithms", or provides seed concepts for graph expansion. Each node is a precisely definable concept (algorithm, theorem, mathematical object) with prerequisite edges.
Use when the user asks what was said earlier in a chat, wants an old decision, exact wording, prior links, or suspects the agent forgot conversation context across Telegram, BlueBubbles, Feishu, ChatGPT, Claude, or other archived chat sources. Search memory_search first, then search the local conversation archive for exact recall.
Full curator pipeline for autonomous learning from quality repositories. Executes: discovery → scoring → ranking → ingest → learn → vault sync. Writes to procedural memory AND Obsidian vault for Graph View visualization and graduation pipeline. Use for: populating procedural memory with domain patterns, first-time domain learning, comprehensive knowledge building. Triggers: /curator full, 'learn patterns from repos', 'build knowledge base'.
Learn patterns from a specific GitHub repository. Clones, analyzes code structure, extracts patterns, populates procedural memory AND syncs to Obsidian vault for Graph View visualization. Use for: targeted learning from known quality repos, quick knowledge acquisition, specific pattern extraction. Triggers: /repo-learn, /curator-repo-learn, 'learn from repo'.
经验检索 Skill，根据关键词搜索项目历史记忆（经验记忆+知识记忆+程序记忆SOP+工具记忆+Auto Memory只读）。触发场景：开始复杂任务前、遇到问题时、需要查找历史解决方案、查找项目理解文档。