OpenBot
CommunityDownloadApp
All plugins

Firecrawl

Web extraction and search

Firecrawl lets agents scrape pages, extract structured data, and search the web through one API. An API key from the Firecrawl dashboard is passed to the local server as one environment variable.

What an agent can do with Firecrawl

  • Turn a page into data

    The firecrawl_scrape tool reads one URL and returns it as Markdown or as structured JSON. The agent can then put prices, specifications or contact details into a table. The firecrawl_parse tool reads PDFs and other documents in the same way.

  • Search the web with sources

    The firecrawl_search tool searches the web and can return the page content with each result. The agent can compare the pages and give the address of each claim. The firecrawl_map tool lists the URLs of a site and does not fetch their content.

  • Crawl a documentation site

    The firecrawl_crawl tool reads many pages of one site and returns their data. For a long job, the agent uses firecrawl_check_crawl_status with the job ID. Use it to collect a full set of docs before the agent writes a summary or a migration plan.

  • Watch a page for changes

    The firecrawl_monitor tools make recurring monitors for a page. Each monitor runs checks and keeps the differences between versions. The agent can read these differences and tell you what changed on a pricing page, a status page or a release notes page.

How to connect Firecrawl to OpenBot

  1. Open Firecrawl in OpenBot

    Press Open in OpenBot on this page, or find Firecrawl in the app's Apps tab. Get the app first if it is not on this computer.

  2. Add your API key

    Press Connect and enter this value. OpenBot keeps it on the computer that runs OpenBot.

    • API key - Firecrawl dashboard → API keys

    Get a key

  3. Give an agent a task

    Ask an agent for work that needs Firecrawl, such as one of the example prompts below. The agent uses the plugin's tools to do it.

Example prompts3

  • FirecrawlExtract the pricing table from this page as structured data.
  • FirecrawlResearch competitor pricing and cite each source page.
  • FirecrawlWhat changed on our changelog page this month?

Apps1

  • Firecrawl

    Page scraping, extraction, and web search, over a local MCP server with a Firecrawl API key.

Information6

Developer
firecrawl.dev
Category
Research
Version
1.0.0
Website
firecrawl.dev
Privacy policy
firecrawl.dev/privacy
Terms
firecrawl.dev/terms

Questions

Is the Firecrawl MCP server official?

Yes. Firecrawl publishes the firecrawl-mcp package in its own GitHub organization, and the repository calls it the official Firecrawl MCP server. It has the MIT license. OpenBot starts this package on the computer that runs OpenBot.

Does the Firecrawl plugin need an API key?

Yes. Make an API key in the Firecrawl dashboard, then type it in the connect dialog in OpenBot. OpenBot gives the key to the local server as the FIRECRAWL_API_KEY environment variable. The plugin catalog holds no secret.

Does Firecrawl cost anything?

Firecrawl has a free plan with 1,000 credits each month and no credit card. A scrape costs 1 credit for each page. A search costs 2 credits for each 10 results. The firecrawl_credit_usage tool tells the agent how many credits remain on your plan.

What data does Firecrawl get?

Firecrawl gets the requests that the agent sends, such as URLs, search queries and extraction schemas. The local server sends them to the Firecrawl API with your key. Your workspaces, conversations and files stay on the computer that runs OpenBot.

Sources

Every statement about Firecrawl comes from the pages below, checked on . They can change after that date.