Mistral Moderation API by Mistral AI

# Fix list: Mistral Moderation API

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/mistral-moderation, the October 2026 research run, assessed 1 October 2026. Grade C, 58.6 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 Mistral Moderation API: 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, 40 out of 100, up to 12 more on the total

Why it scored 40: status.mistral.ai runs on Rootly with 90-day uptime bars per component, but none of its components is for moderation (10 of 20, our call for a status page that doesn't cover the endpoint). The page read "All Systems Operational" and the incident history didn't come through to us, so no readable history (5). Limits are set per workspace tier in the console, and we found no published numbers for moderation (5 of 15). The error glossary says how to resolve each status code, and we didn't confirm a Retry-After header (10 of 15). No SLA found (0). mistral-moderation-2603 is 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, 54 out of 100, up to 8.1 more on the total

Why it scored 54: Plain workspace API keys, revocable in the console, with no endpoint scopes (20). The same key reaches files, fine-tuning, agents, batch jobs and paid models, so it can't be limited to moderation (10 of 20). A jailbreaking category in the classifier, one score rather than a document-aware injection check (12 of 15). Usage per workspace in the console, no per-call log found (5 of 15). security.txt valid, and a trust centre that releases documents on request, with no certification, bug bounty or disclosure policy we could read without JavaScript (7 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.

## 3. Payments & pricing, 40 out of 100, up to 7.5 more on the total

Why it scored 40: No x402, MPP or L402 (0). Listed as free on the API pricing page and the model card (20). The free Experiment plan needs no card, though it needs a phone number (20). A person signs up in a browser and verifies a phone (0).

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.

## 4. Maintenance & community, 33 out of 100, up to 5.9 more on the total

Why it scored 33: Mistral Moderation 2 on 1 March 2026 and the retirement of mistral-moderation-2411 on 31 March 2026, both more than 180 days ago (0). No moderation entries in the last 90 days (0). Dated changelog and docs, with one entry since 3 July (three model deprecations on 29 September), and support through the console (10 of 15). Current official SDKs, mistralai on PyPI and @mistralai/mistralai on npm (15). SDKs generated from the OpenAPI spec, not re-checked this run (8 of 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.

## 5. Agent ergonomics, 75 out of 100, up to 4.1 more on the total

Why it scored 75: Each result is 11 booleans and 11 scores, compact and fixed (20 of 25). The raw endpoint has no category or detail switch, while the moderation_llm_v2 guardrail on conversations and agents takes custom thresholds, ignore_other_categories, an action and block_on_error (10 of 20). The error glossary maps status codes to fixes (15 of 20). Classification has no side effects, but there's no retry guidance we could confirm (15 of 20). Two required fields and official SDKs in Python and TypeScript (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.

## 6. Schema & documentation, 85 out of 100, up to 2.4 more on the total

Why it scored 85: OpenAPI document at docs.mistral.ai/openapi.yaml (25). llms.txt and Markdown pages (10). The guide lists the 11 categories and says to use the raw score or set your own threshold, but doesn't say when the classifier is the wrong tool or which languages it covers (13 of 20). model and input are typed, input takes a string or an array, and the chat endpoint takes typed messages (12 of 15). Python, TypeScript and curl examples, and blocked guardrail calls return 403 with the violated categories, thresholds and scores (13 of 15). Dated model ids, but the changelog's newest moderation entry we found is custom guardrails for agents and conversations, and Moderation 2 (1 March 2026) appears on its model card instead (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.

## 7. Transparency & trust, 83 out of 100, up to 1.5 more on the total

Made of editorial 70, provenance 96.

Why it scored 83: Closed service under commercial terms with a French legal entity, SDKs Apache-2.0 (15). Abuse logs are kept 30 days unless zero retention is bought, and data sent on the free Experiment plan may be used for training, which matters here because moderation is free, while the paid default isn't spelt out (15 of 30). Model lifecycle page with notice periods per stage, and the 2411 retirement was dated (20). EU hosting by default, opt-in regional endpoints and a published subprocessor list (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: mistral.ai, registered 2019-05-15 (7 years) (11 of 15)

## 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 moderation calls fall under the Completion API component on the status page, and the incident history for the last 90 days.
- The rate limits for /v1/moderations on each workspace tier.
- Whether text sent to moderation on the Experiment plan is used for training, or whether moderation is exempt.
- Which certifications Mistral holds. The trust centre needs JavaScript.

## Weaknesses

- No moderation component on the status page and no readable incident history
- No custom topics, blocklists or redaction, and jailbreaking is a single category score
- Supported languages aren't listed
- Data sent on the free Experiment plan may be used for training
- No change to the moderation model since March 2026

## 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 /v1/chat/moderations with the full message list when checking an assistant reply. The raw endpoint has no context
- Read category_scores and set your own threshold per category. The booleans use Mistral's cut-offs
- Pin mistral-moderation-2603. The 2411 model was retired on 31 March 2026
- Move to a paid workspace or zero retention if the text you screen shouldn't train models
- For a Mistral-hosted agent, set the moderation_llm_v2 guardrail with block_on_error true and skip the separate call

## What the review panel asked for

- List supported languages
- Add Moderation 2 to the changelog
- a moderation-only key scope
- a stated training exemption for moderation

## 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.