Integration · Company data

Connect Claude & ChatGPT to Firecrawl

Ask your AI to read a web page, crawl a documentation site or pull prices from a list of product pages. Through your Firecrawl account it can scrape pages into clean markdown, search the web, map a site's URLs, extract structured data and set up monitors that check pages for changes. The AI works with the page content, not a pile of HTML.

Connect Firecrawl free

Free plan, no credit card. Takes about a minute.

Try asking

“Read firecrawl.dev/pricing and summarize it.”
The AI scrapes the page as clean markdown, skipping menus and footers, and summarizes it.
“List every URL on docs.example.com that mentions webhooks.”
It maps the site and returns only the URLs that match the word webhooks, without scraping content.
“Crawl the blog section of example.com, max 50 pages.”
It starts a crawl limited to 50 pages and the paths you name, then fetches the scraped pages when the crawl is done.

What your AI can do

Scrape and search

Scrape one page or a list of pages as markdown, HTML, links or screenshots, and search the web with optional page content.

Crawl, map and extract

Map a site's URLs, crawl many pages with path filters, extract structured data with a prompt or schema, or run a research agent.

Monitors and credits

Create and read page monitors and their past checks, and keep an eye on your credit usage.

“Get the product name and price from these five URLs as JSON.”
It starts an extract job with your prompt and a schema, then returns the structured data.
“Watch our competitor's pricing page every morning.”
It creates a monitor with a name, the page as target and a daily schedule that reports changes.
“How many Firecrawl credits do we have left?”
It reads the team's used and remaining credits for the current billing period.

53 tools for Firecrawl

These tools are switched on when you connect. You can switch any of them off, or require your approval before it runs.

cancel_agentDeletesCancel a agent (job id). Destructive action; may require approval. API summary: Cancel an agent job.
cancel_crawlDeletesStop a running crawl by `crawl_id`. Pages already scraped are kept.
cancel_scrapeMoves moneyCancel a scrape (scrape id). Financial or irreversible action; may require approval. API summary: Cancel a batch scrape job.
create_askChanges dataCreate a new ask. Required: question. API summary: Ask the Firecrawl support agent.
create_developerChanges dataCreate a new developer. Required: query. Search developers first to avoid duplicates. API summary: Search the developer index.
create_docs_searchChanges dataCreate a new docs search. Required: question. API summary: Search Firecrawl docs with citations.
create_interactChanges dataCreate a new interact. API summary: Create an interact session.
create_interact_executeChanges dataCreate a new execute for a interact. Required: code. API summary: Execute code in an interact session.
create_monitorChanges dataCreate a monitor that re-checks `targets` on a `schedule` (cron or plain text) and reports changes; needs a `name`. Each run uses credits.
create_monitor_runChanges dataCreate a new run for a monitor. API summary: Run a monitor.
create_params_previewChanges dataCreate a new params preview. Required: url or prompt. API summary: Preview crawl parameters generated from natural language prompt.
create_scrape_interactChanges dataCreate a new interact for a scrape. Required: code. API summary: Interact with the browser session associated with a scrape job.
delete_interactDeletesPermanently delete a interact by session id. Destructive action; may require approval. API summary: Delete an interact session.
delete_monitorDeletesPermanently delete a monitor by monitor id. Destructive action; may require approval.
delete_scrape_interactsMoves moneyPermanently delete a interact by job id. Financial or irreversible action; may require approval. API summary: Stop the interactive browser session associated with a scrape job.
get_agentGet the status and result of an agent job by `job_id`.
get_agent_snapshotGet one snapshot for a agent by snapshot id. API summary: Get an output snapshot of an agent job.
get_batch_scrapeGet the status and results of a batch scrape job by `scrape_id`.
get_crawlGet the status and scraped pages of a crawl by `crawl_id`.
get_credit_usageGet the team's remaining and used Firecrawl credits for the current billing period.
get_extractGet the status and extracted data of an extract job by `extract_id`.
get_monitorGet one monitor's settings, targets and schedule by `monitor_id`.
get_monitor_checkGet one check for a monitor by check id. Results are paginated. API summary: Get a monitor check.
get_paperGet one paper by paper id. API summary: Inspect or read a paper.
get_the_status_of_a_scrape_jobGet one scrape by job id. API summary: Get the status of a scrape job.
list_activesList actives. API summary: Get all active crawls for the authenticated team.
list_agent_tracesList traces for a agent. Optional filters: live view. API summary: Get the execution trace of an agent job.
list_agentsList agents. Optional filters: before. API summary: List agent runs.
list_crawl_errorsList the pages that failed or were blocked by robots.txt in a crawl, by `crawl_id`.
list_historicalsList historicals. API summary: Get historical credit usage for the authenticated team.
list_historicals_get_historical_token_usageList historicals. API summary: Get historical token usage for the authenticated team (Extract only).
list_interactsList interacts. Optional filters: status. API summary: List interact sessions.
list_monitor_checksList the past checks of a monitor by `monitor_id`, optionally filtered by `status`, to see what changed.
list_monitorsList your page monitors that re-check websites on a schedule for changes.
list_queue_statusesList queue statuses. API summary: Metrics about your team's scrape queue.
list_scrape_errorsList errors for a scrape. API summary: Get the errors of a batch scrape job.
list_tasksList tasks. Optional filters: endpoint. Results are paginated. API summary: List recent API activity.
list_threat_protectionsList threat protections. API summary: Get the team's threat protection policy.
list_token_usagesList token usages. API summary: Get remaining tokens for the authenticated team (Extract only).
map_siteChanges dataList the URLs found on a website starting from `url`, fast and without scraping content. Use `search` to only return URLs matching a word.
scrape_pageChanges dataScrape one web page by `url` and return its content as clean markdown (or other `formats` such as html, links, screenshot). Set `only_main_content` to skip menus and footers.
search_developersSearch developers by query, k, types, repos, sources or skills. Use this to find existing developers before creating a new developer. API summary: Search the developer index.
search_paper_similarsSearch similars for a paper by intent, mode, k, rerank or anchor. Use this to find existing similars before creating a new similar. API summary: Find related papers.
search_papersSearch papers by query, k, authors, categories, from or to. Use this to find existing papers before creating a new paper.
search_webChanges dataSearch the web for `query` and return result links with titles and snippets; add `scrape_options` to also get each result's page content. Use `limit` and `country` to narrow.
start_agentChanges dataStart a research agent that browses the web to answer a `prompt`, optionally limited to `urls` and shaped by a JSON `schema`. Returns a job ID; use get_agent for the result. `max_credits` caps spend.
start_batch_scrapeVisible to othersStart scraping a list of `urls` in one job. Returns a job ID; use get_batch_scrape to fetch the results. Uses credits per page.
start_crawlChanges dataStart crawling a website from `url`, scraping many pages. Limit it with `limit`, `include_paths` and `exclude_paths`. Returns a crawl ID; use get_crawl to fetch results. Uses credits per page.
start_extractChanges dataExtract structured data from one or more `urls` using a plain-language `prompt` and/or a JSON `schema`. Returns a job ID; use get_extract for the results.
submit_feedbackVisible to othersSubmit a feedback (job id). Has effects visible outside this system. API summary: Submit feedback for a search job.
submit_feedback_submit_endpoint_feedbackVisible to othersSubmit a feedback. Has effects visible outside this system. API summary: Submit feedback for a v2 job.
update_monitorChanges dataUpdate an existing monitor identified by monitor id. Only the fields you pass are changed. API summary: Update a monitor.
update_threat_protectionsVisible to othersUpdate an existing threat protection. Has effects visible outside this system. API summary: Update the team's threat protection policy.

Set up in three steps

1
Pick the app
Create a free HeyMyApps account and choose Firecrawl from the app list.
2
Paste your key
Paste your Firecrawl API key. Sign in at firecrawl.dev and open the API Keys page of your dashboard (firecrawl.dev/app/api-keys). Copy your key, which starts with fc-, and paste it here.
3
Add the link to your AI
You get a personal MCP link. Add it to Claude, ChatGPT or Cursor:
1. Click your name, then Settings › Connectors › Add custom connector.
2. Paste your link as the Remote MCP server URL.
3. Switch it on from the + menu in a chat.

Questions

What can the AI do in Firecrawl?

It can scrape single pages or batches of URLs, search the web, map a site's URLs, crawl a site, extract structured data, run a research agent, create and read monitors, and check your credit usage. It can also stop a running crawl.

Does the AI see my Firecrawl API key?

No. You paste your Firecrawl API key into HeyMyApps once. It is encrypted at rest, never shown to the AI and never shown again after you save it. The AI only gets the results of the tools you have switched on.

Can I control what the AI is allowed to do?

Yes. Switch on only the tools you want, for example scraping and search without crawls or monitors. Tools that spend credits on many pages, like crawls, batch scrapes and monitors, can require your approval before they run. Every tool call is logged.

Does it work with ChatGPT?

Yes. In ChatGPT, open Settings › Apps & Connectors › Advanced, turn on Developer mode and add your HeyMyApps link. Developer mode needs a paid ChatGPT plan: Plus, Pro, Business or Enterprise. The same link also works in Claude (Settings › Connectors › Add custom connector), Cursor and other MCP clients.

What does it cost?

HeyMyApps has a free plan, and you do not need a credit card to start. Paid plans are billed per completed task. You connect your own Firecrawl account, and scrapes, crawls and other jobs use your Firecrawl credits.

How do I keep credit use under control?

Crawls and batch scrapes use credits per page, and each monitor run uses credits too. Ask the AI to set a page limit on crawls, use max_credits on agent jobs, and check get_credit_usage when you want to see what is left. You can also require approval before any crawl or monitor starts.