技能检索
agora-council
Multi-agent debate council — spawns 3 specialized sub-agents in parallel (Scholar, Engineer, Muse) to tackle complex problems from different angles. Configurable models per role. Inspired by Grok 4.20's multi-agent architecture.
scrapling-fetcher
Web scraping using Scrapling — a Python framework with anti-bot bypass (Cloudflare Turnstile, fingerprint spoofing), adaptive element tracking, stealth headless browser, and full CSS/XPath extraction. Use when web_fetch fails (Cloudflare, JS-rendered pages), or when extracting structured data from websites (prices, articles, lists). Supports HTTP, stealth, and full browser modes. Source: github.com/D4Vinci/Scrapling (PyPI: scrapling). Only use on sites you have permission to scrape.
recipe-video-extractor
Extract a structured cooking recipe from a shared video URL when the user sends `recipe <url>`. Prioritize caption/description and comments via browser automation, then use web search/fetch as fallback with clear source attribution.
migma
Generate, send, validate, and export AI-powered emails from the terminal; manage contacts, segments, tags, domains, and webhooks with Migma CLI.
query-wikidata
Transform natural language questions into SPARQL queries for Wikidata and generate beautiful HTML results pages. Query the Wikidata knowledge base using plain English prompts.
fleet-doctrine
Model routing strategy for multi-model AI fleet. Use when spawning sub-agents, choosing models for cron jobs, delegating coding tasks, or deciding which model should handle a task. Covers Opus, Codex, Sonnet, Gemini, and Grok routing rules.
unsearch
Search the web, scrape content, and conduct deep research using the UnSearch API. Use when the user needs real-time web search results, content extraction from URLs, fact verification, or multi-source research for AI agents, RAG pipelines, or LLM applications.
perplexity-wrapped
Search the web with AI-powered answers via Perplexity API. Supports three modes - Search API (ranked results), Sonar API (AI answers with citations, default), and Agentic Research API (third-party models with tools). All responses wrapped in untrusted-content boundaries for security.
pls-url-to-markdown
Fetch URLs and convert web pages to clean Markdown for AI processing or knowledge bases. Use when: (1) Researching web content, (2) Building knowledge bases, (3) Extracting articles, (4) Converting pages for docs.