Firecrawl MCP by Firecrawl (Mendable)

# Fix list: Firecrawl MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/firecrawl-mcp, the October 2026 research run, assessed 1 October 2026. Grade BB, 75.5 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on Firecrawl MCP: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Payments & pricing, 50 out of 100, up to 6.3 more on the total

Why it scored 50: No x402, MPP or L402 in the docs index, README or pricing page (0). Plans are public with per-endpoint credit costs (1 a page, search 2 per 10 results, +4 for JSON formats), and $5 top-ups only on paid plans, so between plan-only and per-unit (15 of 20). Free plan with 1,000 credits a month, no card (20). The hosted keyless endpoint gives scrape, search and parse with no signup, but crawl, map and agent need a key (15 of 20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 2. Reliability, 70 out of 100, up to 6 more on the total

Why it scored 70: Better Stack status page at status.firecrawl.dev with history (20). Four incidents since 1 July 2026, all partial degradations of api.firecrawl.dev, the longest 56 minutes on /interact on 20 July and the others 7 to 46 minutes (20). Rate limits published per plan and endpoint, from 10 scrapes a minute on Free to 10,000 on Scale, plus concurrent browsers and a per-IP daily cap for keyless use (15). Exceeding a limit returns 429, but we found no Retry-After or backoff guidance in the rate-limit page or the MCP README (5 of 15). SLA on Enterprise only, per the pricing page, with no terms published (0). GA (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 3. Security & auth, 67 out of 100, up to 5.8 more on the total

Why it scored 67: OAuth on the hosted endpoint `/v2/mcp-oauth`, a separate OAuth identity for the search-only endpoint, and plain revocable keys as `Authorization: Bearer`. Keys locked to endpoints and formats are Enterprise only (25 of 30). The README says never to put a key in the server URL. The legacy key-in-path routes still exist but aren't documented as an option, so no deduction. Keyless and search-only surfaces limit what a session can do, write tools carry destructiveHint, and Alexandria provider terms need explicit user consent and `confirmed: true` (18 of 20). Retained Alexandria results tell the model that source content is data, not instructions, and Enterprise Threat Protection blocks risky URLs, off by default. Nothing covers ordinary scraped pages (5 of 15). firecrawl_credit_usage shows spend, but we found no per-call log for operators (5 of 15). SOC 2 Type II and a valid security.txt per the 26 September check. No bug bounty found (14 of 20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 4. Transparency & trust, 76 out of 100, up to 2.1 more on the total

Made of editorial 60, provenance 92.

Why it scored 76: MIT (30). Zero data retention on Enterprise and a DPA from Standard up, but the privacy policy dates from 26 December 2024, keeps personal data until you ask for deletion and gives no retention period for scraped content (12 of 30). The v1 endpoints are kept as legacy alongside v2 and the pricing change of 4 September 2026 was dated, but we found no deprecation policy (8 of 20). Data stored in the United States, and Stripe, PostHog, Crisp and Vercel Analytics are named, with no full subprocessor list (10 of 20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

Provenance checks not met in full (half of this category, computed from checked facts):

- Domain age: firecrawl.dev, registered 2024-04-08 (2 years) (7 of 15)

## 5. Schema & documentation, 88 out of 100, up to 2 more on the total

Why it scored 88: Public OpenAPI at docs.firecrawl.dev/api-reference/v2-openapi.json (and v1), and zod schemas on every MCP tool (25). llms.txt and Markdown docs (10). The README and server instructions say when not to use a tool, for example to look elsewhere when a browser session must be driven step by step across many calls (18 of 20). Typed inputs with enums, but open issue #325 counts 132 parameters with no description and #373 reports a published schema that disagrees with the API (11 of 15). Examples throughout, and keyless failures return structured recovery payloads with `next_actions` and `signup_url`. We found no error-code reference in the docs index (12 of 15). A CHANGELOG and firecrawl.dev/changelog, but the CHANGELOG skips 3.22 to 3.24 and still lists 3.25.0 as unreleased (12 of 15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 6. Agent ergonomics, 92 out of 100, up to 1.3 more on the total

Why it scored 92: 26 tools in the full profile, 3 on the keyless endpoint and 8 on the search-only endpoint, and env flags drop the feedback tools (23 of 25, the 11 to 30 band plus 8 for fixed subsets). Format selection, main-content filtering, page limits on map and crawl, and results over 20,000 estimated tokens handed off to retained storage instead of inline (20). Recovery payloads with next actions, but schema mismatches reported in open issues (17 of 20). readOnlyHint or destructiveHint on all 30 annotated tool definitions, with monitor delete and interact stop marked destructive. No idempotency keys for crawl or agent jobs (17 of 20). One required argument on most tools, and official SDKs in Python and Node (15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 7. Maintenance & community, 87 out of 100, up to 1.1 more on the total

Why it scored 87: firecrawl-mcp 3.27.2 on npm on 2026-10-01 (30). More than 20 version bumps since 8 July (20). 83 open issues, many listing spam, and reported bugs from July and August (#325, #357, #373) show no fix in the repository yet. Reply times weren't visible (12 of 25). io.github.firecrawl/firecrawl-mcp-server in the official registry under a GitHub-verified namespace (15). CI builds and tests on push and pull request, with npm trusted publishing (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- Whether the API sends Retry-After on 429; the docs we read don't say.
- Reply times on the 83 open issues, which the issues page didn't show us.
- Retention of scraped content outside Enterprise zero data retention; the privacy policy doesn't say.

## Weaknesses

- No x402 or other machine payment
- No prompt-injection guidance for scraped pages outside retained Alexandria results
- 403 and 404 responses cost a credit
- Open issues report 132 undescribed parameters and a schema mismatch
- Privacy policy last revised December 2024, with no retention period for scraped content

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- Use `firecrawl_map` to list URLs, then scrape the few you need instead of crawling a whole site
- On the keyless endpoint only scrape, search and parse work. A 429 carries a `signup_url` for the human
- Ask for markdown only. JSON, question and highlight formats add 4 credits a page
- Poll `firecrawl_check_crawl_status` rather than waiting on a long crawl inside one call
- Check the URL before scraping. Dead pages still cost a credit

## What the review panel asked for

- Publish the keyless cap
- Add per-call x402
- Retry-After on 429
- Fix the schema mismatch
- a changelog entry per published version
- itemised pricing changes
- Describe the 132 parameters
- Publish an error-code reference
- Document Retry-After
- Job idempotency keys
- injection marking everywhere
- retire key-in-path routes
- itemise the September pricing change
- fix schema mismatches
- describe every parameter

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.