MewCP Firecrawl MCP

Turn any website into clean, AI-ready data.

A Model Context Protocol (MCP) server that exposes Firecrawl's API for scraping, crawling, mapping, searching, parsing documents, browser automation, and academic research.

The Firecrawl MCP Server provides powerful web data extraction and research capabilities:

- Scrape individual pages or crawl entire websites into markdown, HTML, JSON, and more
- Search the web, map site structures, and run autonomous agent-based data extraction
- Automate browsers with code or natural language, parse documents, and search academic papers and GitHub

- AI assistants that need to fetch and process live web content
- Automating structured data extraction and research pipelines
- Building competitive intelligence, literature review, and site auditing workflows

Scrapes a single URL and returns its content in the requested formats. Returns the page as markdown, HTML, screenshot, links, or a summary. For public document URLs (PDF, DOCX) Firecrawl auto-detects and parses them. The response includesdata.metadata.scrapeIdwhich can be passed tobrowser_interactto continue interacting with the same live browser session.

{ markdown: string | null; summary: string | null; html: string | null; rawHtml: string | null; screenshot: string | null; links: string[] | null; metadata: { title: string | null; description: string | null; language: string | null; sourceURL: string | null; url: string | null; keywords: string | null; statusCode: number | null; contentType: string | null; error: string | null; scrapeId: string | null; } | null; warning: string | null; }

Starts an async batch scrape job for a list of URLs. Returns a job ID immediately. Useget_batch_scrape_statusto poll for completion and retrieve scraped content. Ideal for scraping 5–1000 URLs in parallel without blocking.

{ id: string; url: string | null; invalidURLs: string[] | null; }

Polls the status of a batch scrape job started bybatch_scrape_urls. Returns status (scraping/completed/failed), progress counters, and scraped pages when done. Ifdata.nextis present in the response, call again with the same job_id to get the next page of results.

- job_id (string, required) — Batch scrape job ID returned by batch_scrape_urls.

{ status: string; total: number | null; completed: number | null; creditsUsed: number | null; expiresAt: string | null; next: string | null; data: { markdown: string | null; summary: string | null; html: string | null; rawHtml: string | null; screenshot: string | null; links: string[] | null; metadata: { title: string | null; description: string | null; language: string | null; sourceURL: string | null; url: string | null; keywords: string | null; statusCode: number | null; contentType: string | null; error: string | null; scrapeId: string | null; } | null; warning: string | null; }[] | null; }

DESTRUCTIVE — REQUIRES EXPLICIT USER CONFIRMATION BEFORE CALLING. Stops a running batch scrape job. All in-progress scraping is terminated and any unfinished results are permanently lost — this cannot be undone. NEVER call this tool autonomously or as part of an automated flow. You MUST stop, tell the user exactly which batch scrape job will be cancelled and that unfinished results will be permanently lost, and wait for their explicit written confirmation before proceeding.

- job_id (string, required) — Batch scrape job ID to cancel.

Starts an async crawl job from a seed URL, following internal links up to the specified depth and page limit. Returns a job ID immediately. Useget_crawl_statusto poll for progress and results. Useinclude_paths/exclude_pathsregex patterns to control which URLs are visited. Ideal for extracting all content from a site, documentation, or blog.

Polls the status of a crawl job started bycrawl_url. Returns status (scraping/completed/failed/cancelled), progress counters, and crawled pages. Ifdata.nextis present, call again to retrieve the next page of results.

- job_id (string, required) — Crawl job ID returned by crawl_url.

{ status: string; total: number | null; completed: number | null; creditsUsed: number | null; expiresAt: string | null; createdAt: string | null; completedAt: string | null; duration: number | null; next: string | null; data: { markdown: string | null; summary: string | null; html: string | null; rawHtml: string | null; screenshot: string | null; links: string[] | null; metadata: { title: string | null; description: string | null; language: string | null; sourceURL: string | null; url: string | null; keywords: string | null; statusCode: number | null; contentType: string | null; error: string | null; scrapeId: string | null; } | null; warning: string | null; }[] | null; }

DESTRUCTIVE — REQUIRES EXPLICIT USER CONFIRMATION BEFORE CALLING. Stops a running crawl job. All in-progress crawling is terminated and any unfinished pages are permanently lost — this cannot be undone. NEVER call this tool autonomously or as part of an automated flow. You MUST stop, tell the user which crawl job will be cancelled and that unfinished pages will be permanently lost, and wait for their explicit written confirmation before proceeding.

- job_id (string, required) — Crawl job ID to cancel.

Discovers all URLs on a website without scraping their content. Returns a list of links with title and description. Use beforecrawl_urlto understand site structure, or passsearchto filter URLs by relevance to a topic. Much faster and cheaper than crawling when you only need the URL list.

- url (string, required) — Root URL of the site to map. - search (string, optional) — Filter and rank URLs by relevance to this search query. - sitemap (string, optional, default: "include") — 'include' (sitemap + crawl), 'skip' (crawl only), 'only' (sitemap only). - include_subdomains (bool, optional, default: true) — Include URLs from subdomains of the root URL. - ignore_query_parameters (bool, optional, default: true) — Deduplicate URLs that differ only in query parameters. - ignore_cache (bool, optional, default: false) — Bypass sitemap cache to get the freshest URL list. - limit (int, optional, default: 5000) — Maximum number of URLs to return (1–100000). - country (string, optional) — ISO 3166-1 alpha-2 country code for geo-targeting (e.g. 'US', 'DE').

{ links: { url: string; title: string | null; description: string | null; }[]; }

Searches the web and optionally scrapes the full content of each result. Returns web pages, images, or news depending onsources. Setscrape_formatsto ['markdown'] to get full page content alongside each result — omit to get only title, description, and URL. Supports operator syntax: site:, filetype:, intitle:, -exclude, "exact phrase".

- query (string, required) — Search query. Supports operators: site:domain.com, filetype:pdf, intitle:keyword, -exclude, "exact phrase", related:domain.com. - limit (int, optional, default: 10) — Number of results to return (1–100). - sources (list[string], optional, default: ["web"]) — Result types to return: 'web', 'images', 'news'. Combine as needed. - categories (list[string], optional) — Filter to specific result categories: 'github', 'research', 'pdf'. - country (string, optional) — ISO country code for geo-targeted results (e.g. 'US', 'DE', 'JP'). Default: US. - location (string, optional) — City/region for geo-targeted results (e.g. 'San Francisco,California,United States'). - tbs (string, optional) — Time-based filter: 'qdr:d' (past day), 'qdr:w' (past week), 'qdr:m' (past month). - include_domains (list[string], optional) — Restrict results to these domains (mutually exclusive with exclude_domains). - exclude_domains (list[string], optional) — Remove these domains from results (mutually exclusive with include_domains). - scrape_formats (list[string], optional) — If provided, each result page is scraped and content returned in these formats. Omit to return only title/description/URL without scraping. - timeout_ms (int, optional, default: 45000) — Request timeout in milliseconds (1000–300000). Default 45000.

{ results: { web: { title: string | null; description: string | null; url: string | null; markdown: string | null; html: string | null; rawHtml: string | null; category: string | null; }[] | null; images: { title: string | null; imageUrl: string | null; imageWidth: number | null; imageHeight: number | null; url: string | null; position: number | null; }[] | null; news: { title: string | null; snippet: string | null; url: string | null; date: string | null; imageUrl: string | null; position: number | null; markdown: string | null; }[] | null; } | null; warning: string | null; id: string | null; creditsUsed: number | null; }

Parses a local or private document (PDF, DOCX, XLSX, HTML, and more) into clean markdown or structured data. Use when the file is not publicly accessible by URL — for public URLs usescrape_urlinstead. The file must be provided as base64-encoded bytes, making this suitable for workflow chains where a previous step fetches and encodes the file content.

- file_content_b64 (string, required) — Base64-encoded file bytes to parse. - file_name (string, required) — Filename including extension (e.g. 'report.pdf', 'data.docx'). Extension determines parser. - formats (list[string], optional, default: ["markdown"]) — Output formats: markdown, html, rawHtml, links, summary. - only_main_content (bool, optional, default: true) — Strip headers, footers, and decorative content.

{ markdown: string | null; summary: string | null; html: string | null; rawHtml: string | null; links: string[] | null; metadata: { title: string | null; description: string | null; language: string | null; sourceURL: string | null; url: string | null; keywords: string | null; statusCode: number | null; contentType: string | null; error: string | null; scrapeId: string | null; } | null; warning: string | null; }

Starts an autonomous web research agent that searches, navigates, and extracts data based on a natural language prompt. No URLs required — the agent finds them. Useschemato get structured JSON output. Returns a job ID; useget_agent_statusto poll. Usespark-1-mini(default, 60% cheaper) for most tasks;spark-1-profor complex multi-domain research. Setmax_creditsto cap spending — the job fails without charges if the limit is hit.

- prompt (string, required) — Natural language description of the data to find (max 10000 chars). Be specific: 'Find the 5 most-funded AI startups in 2024 with founder names and total funding.' - urls (list[string], optional) — Optional seed URLs to focus the agent. Omit to let the agent search freely. - schema (string, optional) — JSON schema string for structured output. Omit for free-form text. - model (string, optional, default: "spark-1-mini") — 'spark-1-mini' (default, cheaper) or 'spark-1-pro' (higher accuracy). - max_credits (int, optional) — Credit cap for this job (default 2500). Job fails without charges if exceeded.

{ id: string | null; status: string | null; data: object | null; expiresAt: string | null; creditsUsed: number | null; }

Polls the status of an agent job started byrun_agent. Returns status (processing/completed/failed/cancelled), extracted data when done, and credit usage. Poll every 15–30 seconds; jobs typically complete in 1–5 minutes.

- job_id (string, required) — Agent job ID returned by run_agent.

{ id: string | null; status: string | null; data: object | null; // shape matches the schema passed to run_agent expiresAt: string | null; creditsUsed: number | null; }

DESTRUCTIVE — REQUIRES EXPLICIT USER CONFIRMATION BEFORE CALLING. Requests cancellation of a running agent job. Any in-progress reasoning steps complete before the job transitions to cancelled — credits for completed steps may still be charged and cannot be recovered. NEVER call this tool autonomously or as part of an automated flow. You MUST stop, tell the user which agent job will be cancelled and the credit implications, and wait for their explicit written confirmation before proceeding.

- job_id (string, required) — Agent job ID to cancel.

Executes code or a natural language prompt in the live browser session bound to a previous scrape job. Thescrape_idcomes fromdata.metadata.scrapeIdin ascrape_urlresponse. First call creates the browser session at the same page state as the scrape. Subsequent calls on the samescrape_idreuse the live session. Provide eithercode(Playwright/Node/Python/Bash to run) orprompt_text(AI-driven navigation), not both. Returns CDP URL, live view URL, stdout, and AI output. Callbrowser_closewhen done to release the session.

- scrape_id (string, required) — Scrape job ID from data.metadata.scrapeId in a scrape_url response. - code (string, optional) — Code to execute in the browser sandbox (1–100000 chars). Provide this OR prompt_text, not both. - prompt_text (string, optional) — Natural language task for the AI browser agent (1–10000 chars). Provide this OR code, not both. - language (string, optional, default: "node") — Code language when using code: 'node' (default), 'python', or 'bash'. - timeout (int, optional, default: 30) — Execution timeout in seconds (1–300).

{ cdpUrl: string | null; liveViewUrl: string | null; interactiveLiveViewUrl: string | null; output: string | null; // AI response when using prompt_text stdout: string | null; result: string | null; stderr: string | null; exitCode: number | null; killed: boolean | null; }

- scrape_id (string, required) — Scrape job ID whose browser session to close (same ID used in browser_interact).

Searches Firecrawl's academic research index by topic, method, benchmark, or author. Returns ranked papers with paperId, title, abstract, and relevance score. UsepaperIdfrom results to callget_paperorfind_related_papers. Supports filtering by author name substring, category (e.g. 'cs.LG'), and date range.

- query (string, required) — Natural language search query (e.g. 'diffusion models image synthesis'). - k (int, optional, default: 40) — Maximum number of ranked papers to return (1–500). - authors (string, optional) — Filter by author name substring (e.g. 'LeCun'). Comma-separate for multiple. - categories (string, optional) — Filter by paper category (e.g. 'cs.LG', 'cs.CV'). Comma-separate for multiple. - from_date (string, optional) — Inclusive lower bound on paper date in YYYY-MM-DD format (e.g. '2023-01-01'). - to_date (string, optional) — Inclusive upper bound on paper date in YYYY-MM-DD format.

{ results: { paperId: string | null; primaryId: string | null; ids: { arxiv: string[] | null; } | null; title: string | null; abstract: string | null; score: number | null; }[]; }

Retrieves full details for a specific research paper by its ID. Returns title, abstract, authors, categories, and dates. Thepaper_idcan be a canonical paperId (e.g. '2014215642691656232') or a source-prefixed ID (e.g. 'arxiv:2105.05233') fromsearch_papersresults.

- paper_id (string, required) — Paper ID — either canonical paperId or source-prefixed ID like 'arxiv:2105.05233'. - k (int, optional) — Number of related papers to include alongside the paper details.

{ paper: { paperId: string | null; primaryId: string | null; ids: { arxiv: string[] | null; } | null; title: string | null; abstract: string | null; authors: string | null; categories: string[] | null; createdDate: string | null; updateDate: string | null; }; }

Finds papers related to a seed paper, ranked by semantic relevance to an intent. Usemodeto choose expansion strategy: 'similar' (semantically close), 'citers' (papers that cite the seed), 'references' (papers cited by the seed). Returns ranked results with relevance scores. Ideal for literature review workflows: search_papers → find_related_papers → get_paper.

- paper_id (string, required) — Seed paper ID (canonical paperId or 'arxiv:XXXX.XXXXX'). - intent (string, required) — Natural language ranking intent (e.g. 'applications in medical imaging'). - mode (string, optional, default: "similar") — Expansion mode: 'similar' (default), 'citers', or 'references'. - k (int, optional, default: 40) — Maximum number of related papers to return (1–500). - rerank (bool, optional, default: false) — Apply an additional reranking pass over the fused candidate set.

{ results: { paperId: string | null; primaryId: string | null; title: string | null; abstract: string | null; score: number | null; }[]; poolSize: number | null; truncated: boolean | null; }

Searches GitHub issue history, pull requests, discussions, and repository READMEs using natural language. Returns matched content with repository metadata, URLs, and markdown snippets. Useful for researching how a bug was fixed, what a library's maintainers have said, or finding prior art in open source projects.

- query (string, required) — Natural language query (e.g. 'race condition in worker shutdown firecrawl'). - k (int, optional, default: 20) — Maximum number of results to return (1–100).

{ results: { resultType: string | null; // issue | pull_request | repository | discussion repo: string | null; url: string | null; pageType: string | null; number: number | null; title: string | null; snippet: string | null; contentMd: string | null; }[]; }

Every tool returns the same top-level envelope. Onlydatavaries per tool.

// Success { success: true; statusCode: number; retriable: false; retry_after_seconds: null; error: null; data: { ... }; // schema shown per tool above } // Error { success: false; statusCode: number; retriable: boolean; retry_after_seconds: number | null; error: { code: string; // VALIDATION_ERROR | AUTH_ERROR | UPSTREAM_ERROR | SERVER_ERROR message: string; details: object; }; data: null; }

- retriable—truewhen it is safe to retry (rate limit, network error, 503).falsefor validation and auth errors.
- retry_after_seconds— seconds to wait before retrying; present only whenretriableistrueand the upstream specifies a delay.
- error.code— machine-readable string:VALIDATION_ERROR,AUTH_ERROR,UPSTREAM_ERROR,SERVER_ERROR.

All scraping tools accept aformatslist:

- markdown— Clean markdown (default)
- html— Cleaned HTML
- rawHtml— Raw page HTML
- screenshot— Page screenshot as base64
- links— All links found on the page
- summary— AI-generated page summary
- json— Structured JSON extraction
- audio,video,branding,product,menu— Specialised extraction modes

- auto— Automatically selects the best proxy (default)
- basic— Standard proxy for general use
- enhanced— Stealth proxy for bot-protected sites (higher credit cost)

batch_scrape_urls,crawl_url, andrun_agentare asynchronous — they return a job ID immediately:
- Call the tool → receivedata.id
- Poll the matching status tool (get_batch_scrape_status,get_crawl_status,get_agent_status) with the job ID
- Keep polling untilstatusiscompleted,failed, orcancelled
- Ifdata.nextis present in the status response, call again with the same job ID to page through results

Recommended polling interval:every 15–30 seconds. Allow at least 2–3 minutes for crawl and agent jobs.

Use thetbsparameter insearch_webto filter results by recency:

qdr:h — Past hour qdr:d — Past day qdr:w — Past week qdr:m — Past month qdr:y — Past year

get_paperandfind_related_papersaccept two ID formats:

Canonical: 2014215642691656232 Source-prefixed: arxiv:2105.05233

UsepaperIdorprimaryIdfromsearch_papersresults.

- Cause:API key not provided in request headers or incorrect format
- Solution:
- VerifyAuthorization: Bearer YOUR_API_KEYandX-Mewcp-Credential-Id: CREDENTIAL-IDheaders are present
- Check API key is active in your MewCP account

- Cause:API calls have exceeded your request limits
- Solution:
- Check credit usage in your Curious Layer dashboard
- Upgrade to a paid plan or add credits for higher limits
- Contact support for credit adjustments

- Cause:No Firecrawl credential linked to your account
- Solution:
- Go toCredentialsin your MewCP dashboard
- Add your Firecrawl API key
- Retry the request with the correctX-Mewcp-Credential-Idheader

- Cause:JSON payload is invalid or missing required fields
- Solution:
- Validate JSON syntax before sending
- Ensure all required tool parameters are included
- Check parameter types match expected values (e.g.timeout_msmust be 1000–300000)

- Cause:Incorrect server name in the API endpoint
- Solution:
- Verify endpoint format:{server-name}/mcp/{tool-name}
- Use the correct server name from documentation
- Check available servers in your Curious Layer account

- Cause:Upstream Firecrawl API returned an error
- Solution:
- Check Firecrawl service status atFirecrawl Status
- Verify your API key has sufficient credits for the operation
- Review the error message in the response for specific details

- Firecrawl Documentation— Official API reference
- Firecrawl API Reference— Complete endpoint reference
- FastMCP Docs— FastMCP specification
- FastMCP Credentials— FastMCP Credentials package for credential handling

Enable AI agents to get structured data from unstructured web with AgentQL.