oto.cx
firecrawl

firecrawl

firecrawl.

List every URL of a site WITHOUT fetching page content — fast and cheap.

ajouter à otoauth : —prefix : firecrawl_*

setup & usage

⚠ firecrawl api key

create an API key in Firecrawl (it starts with fc-), then paste it into oto.

  • byo-only: no shared oto key, each account/org pays for its own credits
    • ▸ read web pages as clean markdown

    fetches the content of a page (or a whole site) as a human sees it: JavaScript is executed, nav and ads are stripped, and usable markdown remains.

  • "summarize this page" / "extract the pricing from this URL" → firecrawl_scrape
  • "what pages does this site have?" → firecrawl_map (URLs only, fast and cheap — do this BEFORE a crawl)
  • "pull the whole blog of this site" → firecrawl_crawl then firecrawl_crawl_status (asynchronous: the job returns an id, poll until completed)
  • "search the web for X and give me the content of the results" → firecrawl_search with scrape_options={"formats": ["markdown"]}
  • "extract the same set of fields from these 30 pages" → firecrawl_extract (give a schema, not just a prompt) then firecrawl_extract_status
    • · cost and choosing the tool

    every rendered page consumes credits: setting limit on a crawl is the only safeguard (API default: 10,000 pages).

  • stable page → max_age accepts a cached version, much cheaper than a fresh render
  • a single page to extract in structured form → firecrawl_scrape with formats=[{"type": "json", "schema": {...}}]: synchronous, a single call, cheaper than extract
  • site that blocks → proxy="stealth" (more expensive, don't set it by default)
  • cookie wall / "see more" button → actions (click, wait, scroll) before capture
  • just the raw HTML of a URL, no JS rendering? serper_scrape is enough. page behind a login? that's the browser connector.
    • · project scope (#605, 2026-08-29)

    under a project with excluded_url_prefixes, firecrawl_search, firecrawl_map and firecrawl_crawl_status drop the matching results/pages and say so (excluded_by_perimeter); firecrawl_scrape, firecrawl_map, firecrawl_crawl and firecrawl_extract refuse a matching URL (the whole batch is refused for extract). firecrawl_extract_status returns data in the caller's schema, not pages: not filtered. details: docs/projects.md.

    ⚙ multiple comptes

    this connector accepts multiple comptes: each stored credential becomes a named account (one name per compte), at your level, your team's or your org's.

  • the first one doesn't need a name; from the second on, each one carries its own (the first then takes the name "principal", renamable)
  • without specification, the agent takes the only stored compte, otherwise the one marked as default; otherwise the call is refused, naming the available comptes — never a random choice
  • target a compte for a call: _account="<name>" on the tool; list them: oto_identity(op='list', connector='firecrawl') (scope='org' or scope='group' for the org's or team's)
  • set the default: oto_identity(op='set', connector='firecrawl', identity_id='<name>'); rename: op='rename' with new_name
  • ⚠️ no tool walks through the comptes on its own: a total across several comptes is obtained by calling each one (_account) and adding up

    • outils

    chargement…

    utilisation

    claude code — installe le plugin oto — claude plugin marketplace add otomata-tech/oto-plugin — mcp + skill configurés.
    claude.ai — branche mcp.oto.cx/mcp en connecteur custom (guide)
    cli — pipx install oto-cli puis oto firecrawl …
    ← tous les connecteurscatalogue mcp →