Firecrawl MCP integration for AI agents.

Connect AI agents to Firecrawl through 30 structured actions, including batch scrape get, start agent, agent cancel, and batch scrape. Review authentication…

ACTION CONTRACTS
30
AUTHENTICATION
API key
SOURCE STATUS
Provider-backed

CATALOG SOURCE REVIEWED AUGUST 23, 2026 / ACTION NAMES AND SCHEMAS DERIVED FROM WORKING MACHINES PROVIDER SOURCE

CAPABILITY PROFILE

What the Firecrawl integration exposes

The Firecrawl Agent App exposes 30 provider-backed actions for data and developer tools work. Its current contract lets an authorized agent get the current status and paged results of a firecrawl batch scrape job by job id, start a firecrawl agent job for multi-page autonomous browsing and extraction, and cancel a running firecrawl agent job by job id. These operations are called through Working Machines as typed capabilities rather than through browser navigation or copied UI steps.

A connection uses API key. Before execution, the agent can inspect the selected action, its required fields, declared scopes, and expected output contract. Provider credentials remain inside the Working Machines runtime; the calling agent receives the capability and its structured result, not the underlying secret.

For reliable operation, start with the narrowest action that satisfies the task, resolve stable provider identifiers before changing state, and validate the returned object or status after execution. Availability still depends on the connected Firecrawl account, granted provider permissions, workspace policy, region, plan, and upstream API behavior.

VERIFIED ACTION SAMPLE

Real Firecrawl capabilities.

Showing 12 of 30 actions. Risk labels are conservative signals based on operation names, not substitutes for provider documentation or runtime policy.

batch_scrape_getREAD

Get the current status and paged results of a Firecrawl batch scrape job by job ID.

INPUTS: id

start_agentWRITE

Start a Firecrawl agent job for multi-page autonomous browsing and extraction.

INPUTS: prompt / urls / schema / maxCredits / strictConstrainToURLs

agent_cancelHIGH IMPACT

Cancel a running Firecrawl agent job by job ID.

INPUTS: id

batch_scrapeREVIEW

Start a Firecrawl batch scrape job for multiple URLs and return the async job ID.

INPUTS: urls / actions / formats / headers / location

batch_scrape_cancelHIGH IMPACT

Cancel a running Firecrawl batch scrape job by job ID.

INPUTS: id

batch_scrape_get_errorsREAD

Get the failed URLs and robots.txt blocks from a Firecrawl batch scrape job.

INPUTS: id

crawlREVIEW

Start a Firecrawl crawl job with compatibility input fields.

INPUTS: url / prompt / includePaths / excludePaths / maxDepth

crawl_cancelHIGH IMPACT

Cancel a running Firecrawl crawl job by job ID.

INPUTS: id

crawl_deleteHIGH IMPACT

Compatibility alias of crawl_cancel for the crawl delete action name.

INPUTS: id

crawl_getREAD

Get the current status and paged results of a Firecrawl crawl job by job ID.

INPUTS: id

crawl_get_errorsREAD

Get the failed URLs and robots.txt blocks from a Firecrawl crawl job.

INPUTS: id

crawl_list_activeREAD

List the currently active Firecrawl crawl jobs for the authenticated team.

INPUT CONTRACTS

Know what the action needs before it runs.

batch_scrape_get

idREQUIRED
The Firecrawl job ID.

start_agent

promptREQUIRED
The natural-language task for the agent.
urlsOPTIONAL
The URLs that constrain where the agent can start.
schemaOPTIONAL
A loose JSON object.
maxCreditsOPTIONAL
The maximum credits the agent may spend.
strictConstrainToURLsOPTIONAL
Whether the agent must stay strictly within the provided URLs.

agent_cancel

idREQUIRED
The Firecrawl job ID.

batch_scrape

urlsREQUIRED
The URLs to scrape in batch.
actionsOPTIONAL
The browser actions to run before each scrape.
formatsOPTIONAL
The output formats to return.
headersOPTIONAL
Custom HTTP headers to send with the request.
locationOPTIONAL
Location settings for the request.

PROVIDER-SPECIFIC WORKFLOWS

Jobs this Agent App can support

Inspect Batch Scrape Get

Get the current status and paged results of a Firecrawl batch scrape job by job ID. Use this as a bounded discovery step, retain the returned identifier, and avoid expanding the read beyond the task's stated scope.

batch_scrape_get

Control Start Agent

Start a Firecrawl agent job for multi-page autonomous browsing and extraction. Resolve the target first, present material changes for confirmation, and make retries idempotent where the provider supports it.

start_agent

Verify with Batch Scrape Get Errors

Get the failed URLs and robots.txt blocks from a Firecrawl batch scrape job. Compare the returned provider state with the intended outcome and preserve stable IDs or canonical links in the run record.

batch_scrape_get_errors

SAFETY BOUNDARY

Operate Firecrawl with explicit limits

  • Authorize Firecrawl with API key and grant only the provider access required by the selected actions.
  • 5 actions are change-capable by name. Confirm the target identity and material parameters before allowing a write.
  • Treat `agent_cancel`, `batch_scrape_cancel`, and `crawl_cancel` as high-impact operations and require an explicit approval boundary.
  • No provider scope string is declared on the sampled actions. Verify the connected account's actual permissions in Firecrawl rather than assuming unrestricted access.
  • After a call, inspect the structured result and execution record before reporting that the Firecrawl task completed successfully.

CONNECTION MODEL

API key

Firecrawl API key used with the Authorization Bearer header. Create it in Firecrawl API Keys: https://firecrawl.dev/app/api-keys.

Working Machines stores provider credentials behind the execution boundary. An agent can use an authorized connection identity, but catalog discovery alone does not reveal OAuth tokens, API keys, or provider secrets.

READ SIGNALS
17
WRITE SIGNALS
1
HIGH IMPACT
4
REVIEW SIGNALS
8

EVIDENCE AND AVAILABILITY

Provider reference

Action names, input fields, authentication types, and counts on this page are generated from the Working Machines provider catalog. Provider behavior, quotas, object semantics, account eligibility, and regional availability remain governed by Firecrawl.

Official Firecrawl website

ONE CONNECTION. REAL WORK.

Give your agent software it can use.

Connect through MCP or explore the Agent App catalog and choose only the capabilities your workflow needs.

EXPLORE AGENT APPS