How is this different from connecting Firecrawl MCP directly?

Connected directly, an agent sees every tool the server exposes and holds the credential itself. Through TrustGate you choose which tools each agent gets, and every call is logged against the agent that made it.

What if the Firecrawl MCP API changes?

We track the Firecrawl MCP schema, so the tool catalogue stays current. Tools you have already allowed keep working, and anything new stays denied until you enable it.

Can I give different agents different Firecrawl tools?

Yes. Access is set per agent: one can get a handful of read tools, another the full set. You grant the whole MCP or single tools within it.

Where can I see what my agents did in Firecrawl?

Every call is logged against the agent that made it, with the tool it used. Activity can be sent on to your own monitoring.

Do I need to run the Firecrawl MCP server myself?

No. TrustGate connects to the Firecrawl MCP server and sits in front of it. Your agent only talks to your TrustGate URL, never to Firecrawl directly.

Connect Firecrawl MCP to your AI agents, securely.

Q: Is Firecrawl MCP secure?

The server itself exposes 30 tools, and some of them can write to your Firecrawl account. Behind TrustGate an agent only sees the tools you allow, requests and responses are inspected, and every call is logged. Anything outside the allowlist is denied before it reaches Firecrawl.

One secured endpoint for every Firecrawl call your AI agents make, scoped per user, inspected, and logged.

Get started for free

Get a demo

Try Firecrawl MCP now

What your agents can ask once Firecrawl MCP is connected.

“Scrape this page and give me the main content as markdown”
“Search the web for recent coverage of this topic”
“Map the URLs on this documentation site”

“Crawl this section and summarise what it covers”
“Parse this PDF and extract the text”
“Research this question across multiple sources”

Select the tools your AI agents can call

30 tools from the Firecrawl MCP schema.

Showing 30 tools

firecrawl_scrapereadScrapes content from a single URL with JS rendering, markdown or HTML output, main-content filtering and tag inclusion or exclusion.
firecrawl_parsereadParses a document such as a PDF and returns its text content.
firecrawl_extractreadDeprecated compatibility entry point. Use firecrawl_scrape with JSON format per known URL, or firecrawl_agent for multi-source research.
firecrawl_searchreadSearches the web and optionally extracts content from the results, across web and semantic sources.
firecrawl_developer_searchreadSearches developer-focused sources for technical questions.
firecrawl_gov_searchreadSearches government sources.
firecrawl_research_search_githubreadDeprecated compatibility entry point. Use firecrawl_developer_search for GitHub issues, pull requests, and READMEs, plus code documentation, returned as matched passages.
firecrawl_search_feedbackwriteRecords quality feedback for a prior Firecrawl search.
firecrawl_mapreadDiscovers the URLs on a site from its sitemap and HTML links.
firecrawl_crawlhigh riskStarts an asynchronous crawl of many pages from a starting URL, with depth control and path filtering. Consumes credits quickly.
firecrawl_check_crawl_statusreadChecks the status and results of a running crawl job.
firecrawl_agenthigh riskRuns an autonomous research agent across multiple sources when the exact pages are not known.
firecrawl_agent_statusreadChecks the status and output of a running research agent.
firecrawl_research_search_papersreadSearch paper metadata and abstracts with a natural-language query across the indexed corpus of biomedical, life-science, clinical, arXiv and other scientific literature.
firecrawl_research_inspect_paperreadRetrieve canonical metadata for one paper ID, such as an arXiv, PMC, PMID, or DOI identifier. Returns the title, abstract, authors, categories, source IDs, and dates as markdown.
firecrawl_research_related_papersreadFinds related papers through citation-graph expansion from one to ten given papers.
firecrawl_research_read_paperreadRetrieve in-body passages from one paper that are relevant to a specific question.
firecrawl_interacthigh riskOperates a live page with clicks, form fills and navigation for dynamic extraction.
firecrawl_interact_stopwriteEnds a live page interaction session.
firecrawl_find_toolsreadBrowses the Alexandria catalogue of categories, providers and tools for a task or website.
firecrawl_feedbackwriteSends product feedback on search results or endpoint quality.
firecrawl_monitor_createwriteCreate a recurring scrape, crawl, or search monitor that compares each check with its retained predecessor.
firecrawl_monitor_listreadList monitors for the authenticated account with optional pagination controls. Returns one page of monitor records and pagination metadata.
firecrawl_monitor_getreadRetrieve one monitor by ID, including its configuration and current state. This does not run or modify the monitor.
firecrawl_monitor_updatewritePatch an existing monitor by ID, changing its name, status, schedule, targets, goal, judging, webhook, notifications, or retention.
firecrawl_monitor_deletehigh riskPermanently delete a monitor by ID and stop its future schedule. This operation cannot be undone and returns deletion status.
firecrawl_monitor_runwriteQueue an immediate check for a monitor outside its normal schedule. This starts network work for the monitor's configured targets and returns the queued check.
firecrawl_monitor_checksreadList historical checks for a monitor, optionally filtered by status and bounded by a result limit. Returns one page of check summaries and pagination metadata.
firecrawl_monitor_checkreadRetrieves one monitor check and its page-level results, optionally filtered by page status.
firecrawl_credit_usagereadGet the authenticated Firecrawl team's current credit balance or its credit usage over the last 90 days.

What Firecrawl MCP exposes

Firecrawl MCP groups its 30 tools into 7 categories, each with a suggested starting policy.

Category	Covers	Access	Suggested policy
Scrape & Parse	Single-page scraping with JS rendering and tag filtering, and document parsing	Read	Treat scraped page content as untrusted input, never as instructions
Search	Web search with optional content extraction, plus developer and government search	Read / Write	Scraped results can carry prompt injection; keep them out of the instruction path
Crawl & Map	URL discovery from sitemaps and links, asynchronous crawling and crawl status	Read / Write	Crawls consume credits quickly; cap depth and page count
Research	Autonomous multi-source research agents and their status	Read / Write	Agent runs are open-ended; approve and bound them
Interact & Discover	Live page interaction with clicks and form fills, and the Alexandria tool catalogue	Read / Write	Interaction acts on live pages and can execute provider tools; approve each run
Monitoring	Creating, listing, reading, updating, running and deleting monitors, and their check history	Read / Write	Monitors run on a schedule and consume credits; approve creation
Account	The team's credit balance and recent credit usage	Read	Read-only; safe to allow