pi-firecrawl

Firecrawl web scraping tools for Pi coding agent - scrape, crawl, map, and search with Firecrawl API v2

Packages

Package details

extension

Install pi-firecrawl from npm and Pi will load the resources declared by the package manifest.

npm report

$ pi install npm:pi-firecrawl

Package: pi-firecrawl
Version: 0.1.0
Published: Jun 14, 2026
Downloads: not available
Author: fosterg4
License: MIT
Types: extension
Size: 13.6 KB
Dependencies: 0 dependencies · 2 peers

Pi manifest JSON

{
  "extensions": [
    "./extensions"
  ],
  "video": "",
  "image": ""
}

Security note

Pi packages can execute code and influence agent behavior. Review the source before installing third-party packages.

README

pi-firecrawl

Firecrawl web scraping extension for Pi coding agent.

Provides 5 tools and a /firecrawl command for Firecrawl API v2 integration.

Install

pi install npm:pi-firecrawl

Tools

Tool	Description
`firecrawl_scrape`	Scrape a single URL into markdown, HTML, raw HTML, links, screenshots, or JSON
`firecrawl_crawl`	Start a crawl job to discover and scrape multiple pages
`firecrawl_crawl_status`	Check the status and retrieve data of a crawl job
`firecrawl_map`	Discover all URLs on a website using the map endpoint
`firecrawl_search`	Search the web and return results

Commands

/firecrawl — Select which Firecrawl tools to enable/disable

Environment Variables

Variable	Default	Description
`FIRECRAWL_API_KEY`	—	Firecrawl API token (fallback if not set via `/firecrawl key`)
`FIRECRAWL_API_URL`	`https://api.firecrawl.dev`	Base URL for Firecrawl API

API Key Management

You can set the Firecrawl API key from within pi (persisted across sessions):

/firecrawl key fc-your-api-key-here

Check or clear the key:

/firecrawl key          # shows current key status
/firecrawl key --clear   # removes saved key

Key resolution order:

Saved key (via /firecrawl key)
FIRECRAWL_API_KEY env var
token parameter passed per-tool

Usage

# Scrape a URL
firecrawl_scrape url="https://example.com"

# Search the web
firecrawl_search query="pi coding agent"

# Start a crawl
firecrawl_crawl url="https://example.com"

# Check crawl status
firecrawl_crawl_status jobId="<job-id>"