{"slug":"firecrawl","name":"Firecrawl MCP","provider_slug":null,"publisher":"official","publisher_name":"Firecrawl","kind":"service-integration","description":"Choose keyless, OAuth or API-key access to Firecrawl search and extraction tools. Check which operations need an account before connecting your client.","tasks":["Read a known URL as content or structured fields.","Find relevant pages before extracting them.","Run a bounded crawl or an authenticated browser interaction."],"limitations":["Keyless hosted access exposes Search, Scrape and Parse only; other selected tools require authenticated access and may depend on plan or team policy.","The current tool guide deprecates the former Extract MCP tool. Use Scrape with JSON for a known page or Agent when sources must be discovered.","A hosted Parse request cannot directly open a local file. Follow its signed-upload handoff; never assume a local path has already been uploaded.","The hosted service version and target success rate were not measured. Client-discovered schemas determine the arguments for the connected tool surface.","A local MCP process is a separate deployment choice and can still call the cloud API. The profiles on this page cover hosted connections."],"capabilities":["web-search","scraping-extraction","browser-automation"],"connections":[{"id":"hosted-keyless","label":"Try Search, Scrape and Parse without an account","transport":"streamable-http","hosting":"hosted","endpoint":"https://mcp.firecrawl.dev/v2/mcp","package":null,"auth":"No key or account; daily per-IP request and credit caps apply.","prerequisites":["Cursor or another client supporting remote Streamable HTTP.","Use the three documented keyless tools; broader tool access requires sign-in or a key."],"config_label":"Cursor mcp.json (JSON)","config":"{\n  \"mcpServers\": {\n    \"firecrawl\": {\n      \"url\": \"https://mcp.firecrawl.dev/v2/mcp\"\n    }\n  }\n}","source_ids":["keyless","limits","cursor"]},{"id":"hosted-oauth","label":"Sign in to a Firecrawl team","transport":"streamable-http","hosting":"hosted","endpoint":"https://mcp.firecrawl.dev/v2/mcp-oauth","package":null,"auth":"OAuth sign-in in a compatible client; no API key pasted into this profile.","prerequisites":["A Firecrawl account and a client supporting MCP OAuth.","Complete browser sign-in and choose the intended team; available tools follow its plan and policy."],"config_label":"Cursor mcp.json (JSON)","config":"{\n  \"mcpServers\": {\n    \"firecrawl\": {\n      \"url\": \"https://mcp.firecrawl.dev/v2/mcp-oauth\"\n    }\n  }\n}","source_ids":["start","oauth","cursor"]},{"id":"hosted-api-key","label":"Use a Firecrawl API key","transport":"streamable-http","hosting":"hosted","endpoint":"https://mcp.firecrawl.dev/v2/mcp","package":null,"auth":"Authorization bearer header; replace YOUR_FIRECRAWL_API_KEY in your private client configuration.","prerequisites":["A Firecrawl account API key and a client supporting HTTP headers.","Use protected client settings for the real key; do not commit a populated configuration or put the key in the endpoint URL."],"config_label":"Cursor mcp.json (JSON)","config":"{\n  \"mcpServers\": {\n    \"firecrawl\": {\n      \"url\": \"https://mcp.firecrawl.dev/v2/mcp\",\n      \"headers\": {\n        \"Authorization\": \"Bearer YOUR_FIRECRAWL_API_KEY\"\n      }\n    }\n  }\n}","source_ids":["keyless","cursor"]}],"tools":[{"name":"firecrawl_search","description":"Find web results from a query; available in keyless mode.","task":"Find official documentation URLs before choosing a page.","source_ids":["tools"]},{"name":"firecrawl_scrape","description":"Read a known URL; JSON format can extract requested fields. Available in keyless mode.","task":"Read a selected public page and return only its needed content.","source_ids":["tools"]},{"name":"firecrawl_parse","description":"Parse supported files. Hosted local-file input requires an upload handoff.","task":"Convert a document after completing the required upload step.","source_ids":["tools"]},{"name":"firecrawl_map","description":"Discover site URLs with authenticated access.","task":"Identify a site section before starting a bounded extraction.","source_ids":["tools"]},{"name":"firecrawl_crawl","description":"Traverse a site scope with authenticated access; normally waits for a terminal job state.","task":"Collect an explicitly limited documentation section.","source_ids":["tools"]},{"name":"firecrawl_check_crawl_status","description":"Resume a crawl wait using its job ID.","task":"Check an existing job after the original wait times out.","source_ids":["tools"]},{"name":"firecrawl_interact","description":"Operate a live page with authenticated access.","task":"Navigate or fill a form only within the task you authorized.","source_ids":["tools"]},{"name":"firecrawl_interact_stop","description":"Release an Interact session using its scrapeId.","task":"Close the browser session when the requested interaction finishes.","source_ids":["tools"]}],"commercial":{"account_required":false,"product":"Keyless hosted Search, Scrape and Parse; a Firecrawl account/team for authenticated tools.","billing":"Keyless use has daily per-IP caps. Account use consumes plan credits and any option charges; returned target error pages can be billable. Scrape JSON adds extraction credits, and browser sessions use separate time-based billing.","pricing_url":"https://www.firecrawl.dev/pricing"},"sources":[{"id":"start","label":"Firecrawl MCP connection modes","url":"https://docs.firecrawl.dev/mcp-server","claims":["Hosted keyless, OAuth and API-key modes; plan and team scope"],"checked_at":"2026-09-28"},{"id":"keyless","label":"Firecrawl keyless and bearer-key configuration","url":"https://docs.firecrawl.dev/mcp-server/keyless","claims":["Remote endpoint, Cursor examples, keyless tools and bearer header"],"checked_at":"2026-09-28"},{"id":"oauth","label":"Firecrawl OAuth account connection","url":"https://docs.firecrawl.dev/mcp-server/oauth","claims":["OAuth endpoint and client sign-in"],"checked_at":"2026-09-28"},{"id":"tools","label":"Firecrawl current MCP tools and troubleshooting","url":"https://docs.firecrawl.dev/mcp-server/tools","claims":["Tool names, access scope, Extract deprecation, upload and session behavior"],"checked_at":"2026-09-28"},{"id":"limits","label":"Firecrawl keyless and account rate limits","url":"https://docs.firecrawl.dev/rate-limits","claims":["Daily anonymous caps versus team account rate and concurrency limits"],"checked_at":"2026-09-28"},{"id":"billing","label":"Firecrawl credit billing","url":"https://docs.firecrawl.dev/billing","claims":["Base credits, extraction modifiers and failure exceptions"],"checked_at":"2026-09-28"},{"id":"cursor","label":"Cursor MCP configuration and authentication","url":"https://cursor.com/docs/mcp","claims":["mcpServers JSON, HTTP headers, OAuth and configuration locations"],"checked_at":"2026-09-28"},{"id":"repository","label":"Official Firecrawl MCP repository","url":"https://github.com/firecrawl/firecrawl-mcp-server","claims":["Publisher, server implementation and MIT license"],"checked_at":"2026-09-28"},{"id":"pricing","label":"Firecrawl plans and pricing","url":"https://www.firecrawl.dev/pricing","claims":["Account plan offers; consult detailed billing for option charges"],"checked_at":"2026-09-28"}],"checked_at":"2026-09-28","updated_at":"2026-09-28","review_due_at":"2026-10-28","verification":"docs-confirmed","test":null,"status":"published","withdrawal_reason":null,"repository_url":"https://github.com/firecrawl/firecrawl-mcp-server","license":"MIT","editorial":{"seo_title":"Firecrawl MCP: Setup, Authentication, Tools & Credit Limits","seo_description":"Connect Firecrawl MCP with keyless access, OAuth or an API key. Compare available tools, extraction choices, credit limits and common setup errors.","intro":"Start with keyless access for a small Search, Scrape or Parse task. Choose OAuth or an API key when your workflow needs the account tool surface. The connection decision controls which operations your assistant can discover.","selection":{"best_for":["An assistant that needs web search followed by content from a selected URL.","A workflow that chooses explicitly between readable content, schema-guided fields and an authenticated crawl."],"not_for":["A raw residential-proxy endpoint to put in an HTTP client.","Assuming every hosted tool is available anonymously or under every team policy."],"source_ids":["tools","keyless","start"]},"recommended_connection":{"connection_id":"hosted-keyless","reason":"Use this for the three keyless tools when evaluating a bounded public task. Switch to OAuth for interactive account access or the bearer-key profile when your client needs a configured key.","source_ids":["keyless","oauth","cursor"]},"workflow":{"title":"Find a page, choose the output, then validate it","steps":[{"title":"Check the connection and tool list","body":"Save one profile in Cursor, complete any required sign-in, and inspect the discovered tools. A saved JSON file alone does not establish a working connection.","tool_names":[],"source_ids":["keyless","oauth","cursor"]},{"title":"Find the URL only if needed","body":"Use Search when the source is unknown. If you already know the page, proceed to Scrape; a search result is a lead to inspect, not evidence that its content answers the task.","tool_names":["firecrawl_search"],"source_ids":["tools"]},{"title":"Request the useful output","body":"Use Scrape for content from the selected URL. For fixed fields, choose its JSON format and supply the client schema or prompt. Check the returned target content before accepting it; extraction can add credits.","tool_names":["firecrawl_scrape"],"source_ids":["tools","billing"]},{"title":"Keep larger work bounded","body":"Authenticate before a crawl and give it an explicit scope. Resume a timed-out wait with the existing crawl ID. If the task uses Interact, stop its session when finished.","tool_names":["firecrawl_crawl","firecrawl_check_crawl_status","firecrawl_interact_stop"],"source_ids":["tools"]}]},"troubleshooting":[{"symptom":"Only three tools appear","resolution":"Check whether you chose keyless access. For additional tools, change to OAuth or add the bearer header, reconnect, and inspect the tool list again. Team policy can still restrict availability.","source_ids":["keyless","tools"]},{"symptom":"401 after saving the connection","resolution":"For /v2/mcp-oauth, sign in again through the client. For /v2/mcp, check the configured key or switch that entry to the OAuth URL. Start a new client session after the change.","source_ids":["tools"]},{"symptom":"429 or an unexpectedly long wait","resolution":"Check the response and current request/concurrency limits. Reduce concurrent work and wait before retrying. Anonymous caps differ from the account plan; signing in changes the applicable limits.","source_ids":["limits"]}],"faqs":[{"question":"Does Firecrawl MCP require an API key?","answer":"The hosted server has keyless Search, Scrape and Parse. OAuth is another route to authenticated account tools without pasting an API key. A bearer key is available when that fits your client.","source_ids":["start","keyless","oauth"]},{"question":"Should I use the former Extract MCP tool?","answer":"The current guide marks it deprecated. Use Scrape with JSON for a known URL. Firecrawl recommends Agent when source pages must be discovered; inspect the connected tool schema and its account requirements.","source_ids":["tools"]},{"question":"Is a returned page necessarily successful or free to retry?","answer":"No. A returned target error page can still consume credits. Inspect the content before accepting it or retrying, and account for extraction and other optional charges.","source_ids":["billing"]}]},"url":"https://workingproxysites.info/mcp/firecrawl","provider_url":null,"review_due":false,"version":null}