Brave Search API + MCP by Brave

# Fix list: Brave Search API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/brave-search-mcp, the October 2026 research run, assessed 1 October 2026. Grade B, 68.1 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 Brave Search API + 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. Reliability, 55 out of 100, up to 9 more on the total

Why it scored 55: status.brave.app (Better Stack) lists a Brave Search API component, but marks every service "not monitored" and shows no incident history, so half credit (10). The API docs keep their own status log, and its last entry is 27 March 2025 (16 minutes with all keys disabled during a database migration). Nothing readable for the last 90 days (5). Rate limits published per plan, 50 QPS Search, 2 QPS Answers, 100 QPS Spellcheck and Autosuggest, in a 1-second sliding window (15). 429 on overrun, `X-RateLimit-Reset` in seconds and exponential backoff guidance. Searches are reads, so no idempotency keys are needed (15). No SLA on any self-serve plan (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.

## 2. Security & auth, 58 out of 100, up to 7.4 more on the total

Why it scored 58: Plain API keys in the `X-Subscription-Token` header. Several keys per account, revocable from the dashboard, with rotation advice, but no scopes (20). Read-only by nature, with tool allow and deny lists in the MCP server. The HTTP transport binds to 127.0.0.1 and checks Origin (18). Results are untrusted web text and LLM Context returns page content, and we found no prompt-injection guidance in the API docs or the MCP README (0). No per-call audit log found. Rate-limit headers show quota use only (0). SOC 2 Type II attestation, a HackerOne bounty, a pen test in April 2025 and a valid security.txt per the 30 September check (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.

## 3. Schema & documentation, 72 out of 100, up to 4.6 more on the total

Why it scored 72: No public OpenAPI for the keyed API. The x402 proxy publishes an OpenAPI 3.1 document per the 30 September check, and all eight MCP tools carry typed JSON Schema (2020-12 since v2.1.1), so 20 of 25. llms.txt at api-dashboard.search.brave.com (10). Each MCP tool says when to use it, none says when not to, and `brave_summarizer` still tells the model it needs a Pro AI subscription that the current plans don't sell (12). Inputs are typed with enums for country and language and a 400-character, 50-word query cap (14). Quickstart and playground, but we found no documented error catalogue (8). Dated `Api-Version` header versions, but no API changelog. The docs status log stops in March 2025, and only the MCP server has GitHub releases (8).

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.

## 4. Agent ergonomics, 72 out of 100, up to 4.6 more on the total

Why it scored 72: Eight tools, under ten, with `--enabled-tools` and `--disabled-tools` to trim the list. Country and language enums make the web search schema long (22). `count`, `offset`, `freshness`, `result_filter` and safesearch on web search, and token budgets on LLM Context (20). 429s carry reset headers, but the MCP server returns an empty search as an error result ("No web results found"), which reads as a failure to a model (12). Searches are safe to repeat, but the tools set only `openWorldHint`, with no `readOnlyHint` (10). Only `query` is required. No official SDK in any language (8).

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.

## 5. Payments & pricing, 70 out of 100, up to 3.8 more on the total

Why it scored 70: x402 V2 on a Brave-run proxy at search.agent.s.brave.app, not on api.search.brave.com. It sells web, LLM Context, news, video, image and local paths at $0.005 in USDC on Base, and not Answers, Autosuggest or Spellcheck, so partial (30). $5 per 1,000 searches published without login (20). The $5 monthly credit needs a card on file (0). An agent with a wallet can search through the proxy with no account or key (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.

## 6. Transparency & trust, 82 out of 100, up to 1.6 more on the total

Made of editorial 63, provenance 100.

Why it scored 82: The MCP server is MIT. The API is closed with published terms, so between the two bands (20). Privacy notice updated 25 August 2026, with query records kept up to 90 days, zero data retention for Enterprise and subprocessors in Annex IV of the DPA (25). Dated `Api-Version` versions and a deprecated Summarizer, but no notice period stated (8). Brave Software Inc. in the US is named, and subprocessors sit in the DPA annex. We found no data-location statement (10).

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.

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

Why it scored 87: MCP server v2.1.4 on 2026-09-17 (30). Five tagged releases since 3 July, v2.1.0 to v2.1.4 (20). Recent pull requests merge within days, but 21 issues are open and several have no reply, including a JSON-RPC batch amplification report from 14 April 2026 and an HTTP transport authentication report from 4 August 2026 (12). Listed in the official MCP registry as io.github.brave/brave-search-mcp-server, a GitHub-verified namespace (15). CI runs format check, build and tests on every push and pull request, with pinned actions and npm audit fixes (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.

- unchecked: incident history for the Search API in the last 90 days, since neither status source shows any
- unchecked: whether the keyed plans bill failed or empty requests
- Whether `brave_summarizer` still works for keyed users now the Summarizer endpoint is deprecated

## Weaknesses

- The status page marks the Search API "not monitored" and shows no incident history
- A card is required even for the $5 monthly credit
- No public OpenAPI for the keyed API and no API changelog
- No SLA on self-serve plans, and Answers is limited to 2 QPS
- No prompt-injection guidance, though LLM Context returns raw page text

## 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.

- Call /res/v1/llm/context when you need page text rather than links. It saves a separate fetch
- Treat "No web results found" from the MCP server as an empty result, not an outage
- On 429, wait for `X-RateLimit-Reset` seconds. There's no `Retry-After` header
- Skip `brave_summarizer`. The Summarizer is deprecated and the x402 proxy doesn't sell it
- Pin `Api-Version` to a date, since requests without it get the latest version

## What the review panel asked for

- x402 on every endpoint
- Card-free credit
- empty results as data
- retire the summariser tool

## 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.