菜单
Lightweight Firecrawl web scrape/search/map via pi-firecrawl CLI. Use for web scraping, Firecrawl search, page extraction, sitemap URL discovery, or structured page data when MCP tool schemas are too token-heavy.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Obsidian Vault Integration — Automatically save session logs to your Obsidian vault. Enables persistent AI agent memory across pi sessions.
Show this user guide for Pi coding agent. Use when: user asks how to use pi, pi configuration, available skills, extensions, commands, or needs help getting started.
orchestrate comprehensive repository assessment using parallel specialist analysis
assess AI tool adoption and usage patterns in the team
analyze code review participation and quality by developer
orchestrate comprehensive team assessment using parallel specialist analysis
| name | firecrawl-cli |
| description | Lightweight Firecrawl web scrape/search/map via pi-firecrawl CLI. Use for web scraping, Firecrawl search, page extraction, sitemap URL discovery, or structured page data when MCP tool schemas are too token-heavy. |
Use bash CLI. Do not call Firecrawl MCP tools directly unless CLI fails.
pi-firecrawl scrape <url> --format markdown
pi-firecrawl scrape <url> --format json --prompt 'extract prices' --schema '{"type":"object"}'
pi-firecrawl search '<query>' --limit 5
pi-firecrawl search '<query>' --limit 5 --scrape
pi-firecrawl map <url> --search 'webhook docs' --limit 20
pi-firecrawl extract <url> --prompt 'extract product details' --schema '{"type":"object"}'
--format json with schema.--format markdown.search first, then scrape selected URL.scrape --wait 5000, then map --search, then search.--limit and focused prompts.