f5xc-firecrawl
The f5xc-firecrawl plugin provides local self-hosted web scraping
via the open-source firecrawl
engine. No API keys, no subscriptions, no cloud dependency. All
operations run against the local firecrawl instance on localhost:3002
inside the devcontainer.
Installation
Section titled “Installation”/plugin install f5xc-firecrawl@f5xc-salesdemos-marketplaceCommands
Section titled “Commands”/scrape
Section titled “/scrape”Scrape a single URL and extract content as markdown.
/scrape https://docs.example.com/getting-started/scrape https://example.com --format markdown,links --wait 2000/batch-scrape
Section titled “/batch-scrape”Scrape multiple URLs at once.
/batch-scrape https://example.com https://example.org https://example.net/crawl
Section titled “/crawl”Crawl multiple pages from a starting URL.
/crawl https://docs.example.com --limit 20 --depth 2/crawl https://docs.example.com --include /api/* --exclude /blog/*Discover all URLs on a website.
/map https://docs.example.com/map https://docs.example.com --search api --subdomains/search
Section titled “/search”Search the web and optionally scrape results.
/search "firecrawl web scraping" --limit 10/search "AI tools 2026" --scrape --time month/extract
Section titled “/extract”LLM-powered structured data extraction from web pages.
/extract https://example.com "Extract the main heading and any links"/extract https://example.com/pricing --schema '{"plans": [{"name": "string", "price": "string"}]}'/llmstxt
Section titled “/llmstxt”Generate an llms.txt file for a site.
/llmstxt https://docs.example.comSkills
Section titled “Skills”web-scraper
Section titled “web-scraper”Auto-activates when you ask to scrape a URL, crawl a website, map site URLs, search the web, extract structured data, generate llms.txt, batch scrape multiple URLs, or convert a web page to markdown. Delegates immediately to the firecrawl-operator agent.
Agents
Section titled “Agents”firecrawl-operator
Section titled “firecrawl-operator”Autonomous web scraping agent that executes curl + jq sequences
against the local firecrawl API. Supports 11 protocols covering all
v1 endpoints. Read-only agent (no Write, Edit, or Agent tools).
| Protocol | Endpoint | Type |
|---|---|---|
| HEALTH | GET / | Sync |
| SCRAPE | POST /v1/scrape | Sync |
| BATCH_SCRAPE | POST /v1/batch/scrape | Async |
| CRAWL | POST /v1/crawl | Async |
| CRAWL_CANCEL | DELETE /v1/crawl/:id | Sync |
| CRAWL_ACTIVE | GET /v1/crawl/active | Sync |
| CRAWL_ERRORS | GET /v1/crawl/:id/errors | Sync |
| MAP | POST /v1/map | Sync |
| SEARCH | POST /v1/search | Sync |
| EXTRACT | POST /v1/extract | Async |
| LLMSTXT | POST /v1/llmstxt | Async |
Infrastructure
Section titled “Infrastructure”The plugin requires the firecrawl stack running in the devcontainer:
| Component | Port | Purpose |
|---|---|---|
| Firecrawl API | 3002 | All scrape/crawl/map/search/extract endpoints |
| Playwright | 3000 | JavaScript rendering engine |
| Redis | 6379 | Job queue backend |
| PostgreSQL | socket | Crawl/batch job persistence |
| LiteLLM proxy | OPENAI_BASE_URL | LLM backend for extract (optional) |
The stack starts automatically when ENABLE_FIRECRAWL=true (the
default). A SessionStart hook checks that the API is reachable and
warns if the service is down.
Differences from Cloud Firecrawl
Section titled “Differences from Cloud Firecrawl”This plugin uses the self-hosted open-source version:
- No authentication or API keys required for scraping
- No credit limits or rate limiting
- Uses v1 API endpoints (not v2)
- Browser sessions and deep research not available
- Extract uses your own LLM proxy instead of hosted models
- Runs entirely within the local container network