firecrawl@3.41.0

VERIFIED· install only

Publisher claims: "Search, scrape, and interact with the web for AI agents. Turn any URL into clean Markdown or structured JSON, search the live web and specialized indexes, parse documents, crawl and map sites, and watch pages for changes." Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry Docker MCP CatalogPublisher firecrawl · publishing for 637 daysInstalls 154KLast detonated 14 h ago
$docker run -i --rm mcp/firecrawl
WatchReport an issue

Coverage

Bundled scripts ran with no arguments; no agent task scenario for this bundle.
Tested
5 of 5 bundled scripts run with no arguments, with decoy credentials in the environment · every file read, file written and host contacted by what ran
Not tested
use of the skill by an agent

What this means for the label: Verified here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.

Before you give it your token

It will ask for
API_KEYFIRECRAWL_API_KEYFIRECRAWL_MCP_SEARCH_OAUTH_ONLYFIRECRAWL_OAUTH_INTROSPECT_CACHE_TTL_MSFIRECRAWL_OAUTH_INTROSPECT_SECRETFIRECRAWL_OAUTH_ISSUERFIRECRAWL_OAUTH_TOKENMCP_DELEGATED_CREDENTIAL_SECRETMCP_OAUTH_ACCEPT_LEGACY_V2_MCP_AUDConfiguration, not credentials: CLOUD_SERVICE, FASTMCP_ENDPOINT, FASTMCP_HOST, FASTMCP_PORT, FASTMCP_STATELESS, FIRECRAWL_API_URL, FIRECRAWL_INTEGRATION, FIRECRAWL_MCP_ALLOWED_HOSTS, FIRECRAWL_MCP_LIBRARY_SKILLS, FIRECRAWL_MCP_RESOURCE_URL, FIRECRAWL_MCP_SEARCH_ENABLED, FIRECRAWL_MCP_SEARCH_ENDPOINT, FIRECRAWL_MCP_SEARCH_PORT, FIRECRAWL_MCP_SEARCH_RESOURCE_URL, HTTP_STREAMABLE_SERVER, JOB_PATH, SSE_LOCAL
Where it talks to
firecrawl.dev (its own)Also client.example, status.example, greenhouse.io — not the publisher's domain; check what for before connecting.
What we saw when it ran
no credential files read · no network
What we could not test
use of the skill by an agent

Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names client.example, status.example, greenhouse.io as well as its own vendor. The guard injects API_KEY, FIRECRAWL_API_KEY, FIRECRAWL_MCP_SEARCH_OAUTH_ONLY, FIRECRAWL_OAUTH_INTROSPECT_CACHE_TTL_MS, FIRECRAWL_OAUTH_INTROSPECT_SECRET, FIRECRAWL_OAUTH_ISSUER, FIRECRAWL_OAUTH_TOKEN, MCP_DELEGATED_CREDENTIAL_SECRET, MCP_OAUTH_ACCEPT_LEGACY_V2_MCP_AUD only into requests to firecrawl.dev (and its subdomains) and refuses the rest.

$vet-skills run --secrets API_KEY=firecrawl.dev,FIRECRAWL_API_KEY=firecrawl.dev,FIRECRAWL_MCP_SEARCH_OAUTH_ONLY=firecrawl.dev,FIRECRAWL_OAUTH_INTROSPECT_CACHE_TTL_MS=firecrawl.dev,FIRECRAWL_OAUTH_INTROSPECT_SECRET=firecrawl.dev,FIRECRAWL_OAUTH_ISSUER=firecrawl.dev,FIRECRAWL_OAUTH_TOKEN=firecrawl.dev,MCP_DELEGATED_CREDENTIAL_SECRET=firecrawl.dev,MCP_OAUTH_ACCEPT_LEGACY_V2_MCP_AUD=firecrawl.dev -- docker run -i --rm mcp/firecrawl

Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.

What changed since 3.36.0

vet-skills diff firecrawl 3.36.0 3.41.0

This version reaches for something 3.36.0 did not: 4 new hosts (client.example, firecrawl.dev, greenhouse.io, status.example) · 9 new credential variables (API_KEY, FIRECRAWL_API_KEY, FIRECRAWL_MCP_SEARCH_OAUTH_ONLY, FIRECRAWL_OAUTH_INTROSPECT_CACHE_TTL_MS, …). Compared from the two recorded runs, not from source.

New hosts
client.example, firecrawl.dev, greenhouse.io, status.example
New credential variables
API_KEY, FIRECRAWL_API_KEY, FIRECRAWL_MCP_SEARCH_OAUTH_ONLY, FIRECRAWL_OAUTH_INTROSPECT_CACHE_TTL_MS, FIRECRAWL_OAUTH_INTROSPECT_SECRET, FIRECRAWL_OAUTH_ISSUER, FIRECRAWL_OAUTH_TOKEN, MCP_DELEGATED_CREDENTIAL_SECRET, MCP_OAUTH_ACCEPT_LEGACY_V2_MCP_AUD

The guard can hold an update like this until someone approves it (runtime.on_change: hold). A change that only shows with a real credential, or on a path the run did not reach, is not in a trace and so not in this comparison.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; bundled scripts ran with no arguments; no agent task scenario for this bundle) on 2026-11-19 and scored it 85/100 (Verified). In what ran, nothing we could detect went wrong: no undeclared network egress, no credential reads, no persistent instruction writes. What it does with a real credential was not observed; the static findings and the publisher carry the rest of the score. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 1 s · 2 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
0.0sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.1sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.2sprocessexited 2: Two inputs: print the differences; exit code 1 when there are any.
0.2sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.3sprocessexited 2: Two inputs: print the differences; exit code 1 when there are any.
0.3sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.3sprocessbundled script run in the sandbox with no arguments · not named in the instructions
0.5sprocessexited 1: Node.js v22.23.5
0.5sprocess5 of 5 bundled scripts run with no arguments
0.5sprocessscript backend runs setup only
0.5sprocessscript backend runs setup only
0.5sprocessno install scripts to re-run

Static findings

MEDIUM
File type combined with upload action.github/workflows/image.yml:29 · File type combined with upload action — agent may exfiltrate local data filesrun: echo "version=$(node -p "require('./package.json').version")" >> "$GITHUB_OUTPUT"Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
File type combined with upload actionscripts/surface-common.mjs:406 · File type combined with upload action — agent may exfiltrate local data filesconsole.log(formatDiff(differences, inputs));Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
File type combined with upload action.github/workflows/publish.yml:48 · File type combined with upload action — agent may exfiltrate local data filesrun: echo "version=$(node -p "require('./package.json').version")" >> "$GITHUB_OUTPUT"Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
File type combined with upload actionscripts/surface-common.mjs:389 · File type combined with upload action — agent may exfiltrate local data filesif (input.endsWith('.json')) return JSON.parse(readFileSync(input, 'utf8'));Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
HIGH
Access to sensitive file pathsscripts/surface-common.mjs:179 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesenv: { ...process.env, ...BASE_ENV, ...env },Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Access to sensitive file pathstests/mcp-smoke.test.mjs:2319 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files...env,Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Access to sensitive file pathstests/mcp-smoke.test.mjs:106 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files...env,Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
Access to sensitive file pathstests/helpers/exchange-mcp.mjs:57 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files...env,Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05
MEDIUM
File type combined with upload actionsrc/alexandria-output.ts:188 · File type combined with upload action — agent may exfiltrate local data files… full result. To read the rest, follow nextTool: it runs a command over response.json, where call i\'s data is .data.alexandria[i].data, in a virtual Bash; then reuse workspaceId with your own command. Send each Bash call alone. The shell …Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07
MEDIUM
MCP server using network transport (SSE/HTTP)tests/mcp-jobs-tool.test.mjs:44 · MCP server using network transport (SSE/HTTP) — data sent over networkconst instance = createFirecrawlMcpServer({ transport: 'httpStream', ...options, unstable_hooks: { authenticate: async () => session, onToolResult } });Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
AST04
MEDIUM
Access to sensitive file pathstests/integration-tag.test.mjs:83 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system files...env,Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05

Cost to the agent

Instructions loaded
≈36 tok
1 file · 81 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.