Skip to main content

​
Firecrawl

Shanone’s Firecrawl integration gives your agent 22 tools for turning the web into clean, structured data — single-page scraping, full-site crawling, AI-driven extraction, and even a sandboxed browser for interactive pages.

​
Getting Started

1

Get a Firecrawl API key

Create an API key from the Firecrawl dashboard.
2

Add the key in Shanone

Open Integrations in the Shanone dashboard, select Firecrawl, and paste the API key.
3

Retry your request

Once saved, shanone_execute_tool calls for firecrawl_* tools will succeed.

​
Available Tools

Shanone provides 22 tools for Firecrawl, organized into these categories:
firecrawl_scrape, firecrawl_search
firecrawl_crawl, firecrawl_check_crawl_status
firecrawl_extract, firecrawl_map
firecrawl_interact, firecrawl_interact_stop
firecrawl_create_browser, firecrawl_execute_browser, firecrawl_list_browsers, firecrawl_delete_browser
firecrawl_agent, firecrawl_get_job_status
firecrawl_monitor_create, firecrawl_monitor_list, firecrawl_monitor_get, firecrawl_monitor_update, firecrawl_monitor_delete, firecrawl_monitor_run, firecrawl_monitor_check, firecrawl_monitor_checks

​
Common Use Cases

Clean Markdown extraction

Scrape a JS-heavy page and get back clean Markdown instead of raw HTML

Site-wide knowledge base ingestion

Crawl an entire documentation site and feed the resulting pages into a knowledge base

Schema-based data extraction

Extract structured fields (prices, specs, contact info) from a list of URLs using a JSON schema

Price/content change monitoring

Set up a monitor on competitor pages and get notified when pricing or content changes

​
Troubleshooting

These are async operations by default — set wait_for_completion=True, or poll firecrawl_check_crawl_status / firecrawl_get_job_status with the returned job ID.
Check the error message for “rate limit” (back off and retry) vs. “API key” (the connected Firecrawl API key is invalid or expired and needs to be reconnected).
A plain firecrawl_scrape can’t interact with the page — first scrape to get a scrapeId, then use firecrawl_interact with a prompt or code to perform the interaction, and call firecrawl_interact_stop when finished.