Skip to content
actionlake
Start free

Connectors · AI

Firecrawl MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Firecrawl and the rest of your systems. It gives your agent 56 Firecrawl actions (37 reads, 13 writes and 6 that wait for a confirmation) and holds the Firecrawl credential itself, so the agent never does. It paces Firecrawl's limits and records every call.

Connect Firecrawl freeSee pricing

100,000 free calls a month. No card, no email, no password: press Start a free sandbox on the next screen and keep the key it shows you.

claude mcp add --transport http actionlake https://connect.actionlake.com/mcp
the first call
do {  action: "firecrawl.monitors.list",  intent: "check which Firecrawl account is connected"}# List monitors · read · 1 upstream call
  • 56actionsin one signed file
  • 37readlook, never change
  • 13writecreate and update
  • 6wait for a yesdestructive, confirmed first

Set up

Connect Firecrawl in three steps.

  1. 01

    Add actionlake to Claude.

    Run the command once in Claude Code; its consent page can create a free sandbox: 100,000 calls a month, no card. In the Claude app, add a custom connector with the URL below and paste your key on the consent page. Creating a new sandbox from the Claude app is not yet tested by us.

    https://connect.actionlake.com/mcp
  2. 02

    Connect Firecrawl.

    API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead. actionlake stores what you paste write-only, sealed under your tenant's key.

    • api_keyFirecrawl API key (starts with fc-), from https://www.firecrawl.dev/app/api-keys
    account { op: "connections.requirements", service: "firecrawl" }
# free: exactly what to paste, and what is found for you
  3. 03

    Make the first call. Check before a write.

    Ask in plain words; Claude picks the action. Before anything changes, check validates the call and prices it, for free.

    do    { action: "firecrawl.monitors.list" }# list monitors · read · 1 upstream callcheck { action: "firecrawl.agents.start", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent

Firecrawl at a glance

Sign-in
API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead.
You provide
api_key
Limits we pace to
100 calls a minute, per connection
API covered
Firecrawl API v2api.firecrawl.dev/v2
Built from
Firecrawl's documentation, read 22 October 2026docs.firecrawl.devdocs.firecrawl.dev
Service file
Version 1.0.0, signed (Ed25519)
Cached reads
19 of 37 reads, for up to 5 minutes; a write clears what it affects
Shared entities
2, so ask can line Firecrawl up with your other systems
Data
Processed and stored in Sydney, Australia. No US region yet

Capabilities

What an AI agent can do in Firecrawl.

56 actions in 21 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

  • 37readlook, never change
  • 13writecreate and update
  • 6destructivewait for a confirmation

  1. Team

    7

    6 read · 1 write · 0 destructive

    Get remaining credits for the authenticated team · Get historical credit usage for the authenticated team · Get historical token usage for the authenticated team (Extract only) · Metrics about your team's scrape queue

  2. Crawls

    6

    4 read · 1 write · 1 destructive

    Crawl multiple URLs based on options · Get the status of a crawl job · Get all active crawls for the authenticated team · Preview crawl parameters generated from natural language prompt

  3. Monitors

    6

    2 read · 3 write · 1 destructive

    List monitors · Create a monitor · Run a monitor · Get a monitor

  4. Agents

    5

    3 read · 1 write · 1 destructive

    List agent runs · Start an agent task for agentic data extraction · Get the status of an agent job · Get the execution trace of an agent job

  5. Batch scrapes

    4

    2 read · 1 write · 1 destructive

    Scrape multiple URLs and optionally extract information using an LLM · Get the status of a batch scrape job · Get the errors of a batch scrape job · Cancel a batch scrape job

  6. Browser sessions

    4

    1 read · 2 write · 1 destructive

    Create an interact session · List interact sessions · Delete an interact session · Execute code in an interact session

  7. Research papers

    3

    3 read · 0 write · 0 destructive

    Inspect or read a paper · Find related papers · Search papers

  8. Developer search

    2

    2 read · 0 write · 0 destructive

    Search the developer index · Search the developer index

13 more areas · 19 actions
  • Extracts 2
  • Gov search 2
  • Monitor checks 2
  • Scrape 2
  • Scrape interactions 2
  • Support 2
  • Agent snapshots 1
  • Feedback 1
  • Map 1
  • Parse formats 1
  • Search 1
  • Search feedback 1
  • Team activity 1

Top actions

The Firecrawl API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems.

ActionWhat it doesEffectSame name in
firecrawl.monitors.listList monitorsList monitorsreadthis connector
firecrawl.team_activity.listList recent API activityList recent API activityreadthis connector
firecrawl.map.runMap multiple URLs based on optionsMap multiple URLs based on optionsreadthis connector
firecrawl.scrape.runScrape a single URL and optionally extract information using an LLMScrape a single URL and optionally extract information using an LLMreadthis connector
firecrawl.search.runSearch and optionally scrape search resultsSearch and optionally scrape search resultsreadthis connector
firecrawl.agents.listList agent runsList agent runsreadthis connector
firecrawl.agents.startStart an agent task for agentic data extractionStart an agent task for agentic data extractionwritethis connector
firecrawl.crawls.startCrawl multiple URLs based on optionsCrawl multiple URLs based on optionswritethis connector
firecrawl.monitors.runRun a monitorRun a monitorwritethis connector
firecrawl.extracts.startExtract structured data from pages using LLMsExtract structured data from pages using LLMswritethis connector
firecrawl.feedback.createSubmit feedback for a v2 jobSubmit feedback for a v2 jobwritethis connector
firecrawl.monitors.deleteDelete a monitorDelete a monitordestructivethis connector

Try it

Prompts to paste into Claude.

Each one names the Firecrawl actions it will call, so you know what it reads and what it changes before you send it.

  1. Readread

    List monitors in Firecrawl and tell me what stands out.

    callsfirecrawl.monitors.list

  2. Writewrite

    Start an agent task for agentic data extraction in Firecrawl. Show me exactly what you will send before you save it.

    callsfirecrawl.agents.start

Governance for Firecrawl

Destructive actions wait for a yes.

6 Firecrawl actions are destructive. These are all of them:

  • Delete a monitorfirecrawl.monitors.delete
  • Delete an interact sessionfirecrawl.browser_sessions.delete
  • Stop the interactive browser session associated with a scrape jobfirecrawl.scrape_interactions.stop
  • Cancel an agent jobfirecrawl.agents.cancel
  • Cancel a crawl jobfirecrawl.crawls.cancel
  • Cancel a batch scrape jobfirecrawl.batch_scrapes.cancel

A destructive action runs only with confirm: true. Under the human policy the agent cannot confirm for itself: the run stops until a named person approves, in the console or from a signed link in an email. The cap on a key is refused before the call is sent, never after.

dofirecrawl.monitors.deleteconfirm_policy: human

Delete a monitor?

intent: the user asked me to delete a monitor

run suspended · waiting for a named person · key firecrawl-agent cannot confirm itself
In the console, or from an emailed signed link.
a key for this connector
account {  op: "keys.create", name: "firecrawl-agent",  allowed_actions: ["firecrawl.*"], cap_calls: 2000,  confirm_policy: "human_for_writes",  approval_rules: [    { match: "firecrawl.monitors.delete", approver: "[email protected]" },    { match: "firecrawl.browser_sessions.delete", approver: "[email protected]" },    { match: "firecrawl.scrape_interactions.stop", approver: "[email protected]" }  ]}
Illustrative. Scoped to Firecrawl only, capped at 2,000 upstream calls, every write confirmed, and the destructive ones approved by a named person.

Cost

What Firecrawl costs on actionlake.

Every Firecrawl action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Firecrawl refused with 401, 403, 429 or a 5xx are free. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

  • 1upstream call per actionall 56 Firecrawl actions
  • 8actions pageeach page of a long list is one call
  • 19reads can be cacheda cache hit costs nothing
  • 100,000calls a month, freepaid plans from US$49 a month for 250,000

See pricing

Action index

Every Firecrawl action.

56 actions by area: the id your agent calls, what it does, and its effect.

Team7
  • firecrawl.team.get_credit_usageGet remaining credits for the authenticated teamread
  • firecrawl.team.get_historical_credit_usageGet historical credit usage for the authenticated teamread
  • firecrawl.team.get_historical_token_usageGet historical token usage for the authenticated team (Extract only)read
  • firecrawl.team.get_queue_statusMetrics about your team's scrape queueread
  • firecrawl.team.get_threat_protectionGet the team's threat protection policyread
  • firecrawl.team.get_token_usageGet remaining tokens for the authenticated team (Extract only)read
  • firecrawl.team.update_threat_protectionUpdate the team's threat protection policywrite
Crawls6
  • firecrawl.crawls.cancelCancel a crawl jobdestructive
  • firecrawl.crawls.getGet the status of a crawl jobread
  • firecrawl.crawls.get_errorsGet the errors of a crawl jobread
  • firecrawl.crawls.list_activeGet all active crawls for the authenticated teamread
  • firecrawl.crawls.preview_paramsPreview crawl parameters generated from natural language promptread
  • firecrawl.crawls.startCrawl multiple URLs based on optionswrite
Monitors6
  • firecrawl.monitors.createCreate a monitorwrite
  • firecrawl.monitors.deleteDelete a monitordestructive
  • firecrawl.monitors.getGet a monitorread
  • firecrawl.monitors.listList monitorsread
  • firecrawl.monitors.runRun a monitorwrite
  • firecrawl.monitors.updateUpdate a monitorwrite
Agents5
  • firecrawl.agents.cancelCancel an agent jobdestructive
  • firecrawl.agents.getGet the status of an agent jobread
  • firecrawl.agents.get_traceGet the execution trace of an agent jobread
  • firecrawl.agents.listList agent runsread
  • firecrawl.agents.startStart an agent task for agentic data extractionwrite
Batch scrapes4
  • firecrawl.batch_scrapes.cancelCancel a batch scrape jobdestructive
  • firecrawl.batch_scrapes.getGet the status of a batch scrape jobread
  • firecrawl.batch_scrapes.get_errorsGet the errors of a batch scrape jobread
  • firecrawl.batch_scrapes.startScrape multiple URLs and optionally extract information using an LLMwrite
Browser sessions4
  • firecrawl.browser_sessions.createCreate an interact sessionwrite
  • firecrawl.browser_sessions.deleteDelete an interact sessiondestructive
  • firecrawl.browser_sessions.executeExecute code in an interact sessionwrite
  • firecrawl.browser_sessions.listList interact sessionsread
Research papers3
  • firecrawl.research_papers.getInspect or read a paperread
  • firecrawl.research_papers.list_similarFind related papersread
  • firecrawl.research_papers.searchSearch papersread
Developer search2
  • firecrawl.developer_search.runSearch the developer indexread
  • firecrawl.developer_search.run_advancedSearch the developer indexread
Extracts2
  • firecrawl.extracts.getGet the status of an extract jobread
  • firecrawl.extracts.startExtract structured data from pages using LLMswrite
Gov search2
  • firecrawl.gov_search.runSearch the Government Indexread
  • firecrawl.gov_search.run_advancedSearch the Government Indexread
Monitor checks2
  • firecrawl.monitor_checks.getGet a monitor checkread
  • firecrawl.monitor_checks.listList monitor checksread
Scrape2
  • firecrawl.scrape.getGet the status of a scrape jobread
  • firecrawl.scrape.runScrape a single URL and optionally extract information using an LLMread
Scrape interactions2
  • firecrawl.scrape_interactions.executeInteract with the browser session associated with a scrape jobwrite
  • firecrawl.scrape_interactions.stopStop the interactive browser session associated with a scrape jobdestructive
Support2
  • firecrawl.support.askAsk the Firecrawl support agentread
  • firecrawl.support.search_docsSearch Firecrawl docs with citationsread
Agent snapshots1
  • firecrawl.agent_snapshots.getGet an output snapshot of an agent jobread
Feedback1
  • firecrawl.feedback.createSubmit feedback for a v2 jobwrite
Map1
  • firecrawl.map.runMap multiple URLs based on optionsread
Parse formats1
  • firecrawl.parse_formats.listList file formats supported by /parseread
Search1
  • firecrawl.search.runSearch and optionally scrape search resultsread
Search feedback1
  • firecrawl.search_feedback.createSubmit feedback for a search jobwrite
Team activity1
  • firecrawl.team_activity.listList recent API activityread

Questions

The Firecrawl MCP server, answered.

Is there a Firecrawl MCP server for Claude?

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 56 Firecrawl actions across team, crawls, monitors and agents, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Firecrawl.

What do I need to connect Firecrawl?

Firecrawl API key (starts with fc-), from https://www.firecrawl.dev/app/api-keys. The free call account { op: "connections.requirements", service: "firecrawl" } lists exactly what to paste.

Can an AI agent delete or change things in Firecrawl?

Only if its key allows it, and only with an explicit confirmation. 6 Firecrawl actions are destructive, including delete a monitor, delete an interact session and stop the interactive browser session associated with a scrape job. Each runs only with confirm: true. Under the human confirm policy, or an approval rule on the key, the run waits until a named person approves it.

Will an AI agent use up my Firecrawl API limit?

actionlake paces every call to the limits in the Firecrawl service file: 100 calls a minute per connection. Repeated reads are answered from cache (19 Firecrawl actions can be), which never reaches Firecrawl.

What does a Firecrawl action cost?

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (8 Firecrawl actions page). Cache hits, searches, dry runs and calls Firecrawl refused with 401, 403, 429 or a 5xx are free. The Free plan includes 100,000 calls a month with no card. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

Which Firecrawl API does it cover?

The Firecrawl API v2 at api.firecrawl.dev/v2. It is built from Firecrawl's documentation as read on 22 October 2026. 56 actions in 21 areas, each one in the signed service file (version 1.0.0).

Does actionlake keep my Firecrawl data, and where?

actionlake processes and stores data in Sydney, Australia; there is no US region yet. Reads may be cached for up to 5 minutes, and a write clears the reads it affects. Each run's request and response is kept, sealed, for your plan's trace window: 1 day on Free, 3 days on Starter, 7 days on Pro, 14 days on Team, 30 days on Business and 30 days on Enterprise. Your Firecrawl credential is sealed under your tenant's key, never logged and never returned. Your traffic is not used to train models.

Which AI clients can use it?

Claude Code, tested by us end to end, and the Claude app as a custom connector, tested by us with an existing key. Other MCP clients that support remote servers over Streamable HTTP (MCP revisions 2025-06-18 and 2025-11-25) and OAuth dynamic client registration, or a bearer header, should connect the same way; we have not tested them. Or plain REST, under the same action names.

More connectors like Firecrawl

Joins with, on shared entities

Browse

actionlake is independent. It is not made or endorsed by Firecrawl; the name is the vendor's own. Counts are from the Firecrawl service file, version 1.0.0.

Give your agent Firecrawl today.

Connect Firecrawl freeSee pricing

Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.