Firecrawl

Enables AI agents to scrape and crawl websites through the Model Context Protocol using Firecrawl. Converts web pages into clean, LLM-friendly markdown format, handling JavaScript rendering, pagination, and anti-bot measures. Ideal for extracting structured content from websites at scale.

Package

Registry	npm (stdio)
Package	`firecrawl-mcp@3.33.4`
Integrity	`sha512-A6FF9n5Tqwr014svYil+dp1tHJEYZl5bUVhdIF6aTMnM/zjNzyHq9iP2IJbh9aosgBmD+Ti86bGawPjGgslOSQ==`
Lockfile	package-lock.json `sha512-Gicf8Zw5cdV5jMoQMLbqsQuMlgIjjYWyKFbQjVjC1E/mOrlwD0zWVgdNFIDBCgIvRF6FcqtLFWk/VyxhHT83Ng==`
Entry	`dist/index.js`
Install scripts	none allowed · not run: [email protected]

Configuration

Key	Name	Kind	Required	Secret	Format	Default / choices
`FIRECRAWL_API_KEY`	Firecrawl API Key Firecrawl API key	env	yes	yes	string

Tools (26)

From the test run on 2026-10-23 (darwin-arm64, linux-x64, Node 24) — starts without credentials.

`firecrawl_agent`	params: effort, exchange, maxCredits, mode, prompt*, schema, strictConstrainToURLs, threadId, urls
`firecrawl_agent_status`	params: id*
`firecrawl_check_crawl_status`	params: id*
`firecrawl_crawl`	params: allowExternalLinks, allowSubdomains, crawlEntireDomain, deduplicateSimilarURLs, delay, excludePaths, ignoreQueryParameters, includePaths, limit, maxConcurrency, maxDiscoveryDepth, prompt, scrapeOptions, sitemap, url*, webhook, webhookHeaders
`firecrawl_credit_usage`	params: byApiKey, view
`firecrawl_developer_search`	params: k, query*, skills
`firecrawl_find_tools`	Browse Alexandria data providers and workflows or read a selected contract. Alexandria covers companies, people, jobs, finance and filings, public records and government spending, real estate, places and restaurants, retail and prices, package registries and developer data, news, research, and more: typed, sourced records through published contracts. Prefer normal firecrawl_search for a data task; it already returns matching providers. Use this tool when the contract you need was not returned in full, to browse a category when search found nothing, or before scraping the same fields from several pages. Discovery is free. Use query for semantic discovery or urls to find providers for a website. With no arguments, browse categories, then providers and tools. Contracts describe the inputs and outputs for execution through firecrawl_scrape. nextTool identifies further discovery or pagination. Discovery does not execute providers. Use firecrawl_search when you also need web results. params: capabilities, categories, expand, groups, includeCosts, level, limit, offset, providers, query, urls
`firecrawl_gov_search`	params: k, query*
`firecrawl_interact`	params: code, language, prompt, scrapeId, scrapeOptions, timeout, url
`firecrawl_interact_stop`	params: scrapeId*
`firecrawl_map`	params: ignoreQueryParameters, includeSubdomains, limit, search, sitemap, url*
`firecrawl_monitor_check`	params: checkId, id, limit, pageStatus, skip
`firecrawl_monitor_checks`	params: id*, limit, offset, status
`firecrawl_monitor_create`	params: body, email, excludeDomains, goal, includeDiffs, includeDomains, maxResults, name, page, pages, queries, scheduleText, searchWindow, timezone, webhookUrl
`firecrawl_monitor_delete`	params: id*
`firecrawl_monitor_get`	params: id*
`firecrawl_monitor_list`	params: limit, offset
`firecrawl_monitor_run`	params: id*
`firecrawl_monitor_update`	params: body, id
`firecrawl_parse`	params: contentType, excludeTags, filePath*, formats, includeTags, jsonOptions, maxAge, onlyMainContent, parsers, pdfOptions, proxy, queryOptions, redactPII, removeBase64Images, skipTlsVerification, storeInCache, zeroDataRetention
`firecrawl_research_inspect_paper`	params: paperId*
`firecrawl_research_read_paper`	params: k, paperId, question
`firecrawl_research_related_papers`	params: intent, k, mode, rerank, seed_ids
`firecrawl_research_search_papers`	params: authors, categories, from, k, query*, to
`firecrawl_scrape`	params: actions, alexandria, domainTools, excludeTags, formats, includeTags, jsonOptions, location, lockdown, maxAge, mobile, onlyMainContent, parsers, pdfOptions, profile, proxy, queryOptions, redactPII, removeBase64Images, requestId, screenshotOptions, skipTlsVerification, storeInCache, timeout, toolDetail, url, waitFor, zeroDataRetention
`firecrawl_search`	params: categories, domainTools, enterprise, excludeDomains, filter, highlights, includeDomains, limit, location, query*, scrapeOptions, sources, tbs, toolDetail

Security Assessment

Level	caution
Permissions	read:web_content, write:firecrawl_monitors, interact:remote_browser
Data Exposure	content
Requires Supervisor	No
Risk Notes	Scrapes, crawls and searches via the Firecrawl cloud, which receives the requested URLs, queries and page content. Version 3.x can also create, update and permanently delete scheduled monitors on the Firecrawl account (optionally with webhooks) and drive a live remote browser that clicks, fills forms or runs Bash/Python/Node code in Firecrawl's sandbox.
Review	pass · community · 2026-10-21
Runtime	engines.node >=22.0.0 · native modules: no · 50.6 MB installed

When to Use

Use	When the agent needs to scrape web pages and convert them to clean markdown for processing.
Do not use	When the task must not send URLs or page content to the Firecrawl cloud, or when no change to Firecrawl monitors should be possible.
Input / Output	url, text → text, markdown, json

io.github.firecrawl/firecrawl-mcp-server · v3.33.4 · Category: content · Free tier: Yes · Firecrawl offers a free tier with limited pages; paid plans for higher volume. · Source · Website