OpenAI Moderation API by OpenAI

# Fix list: OpenAI Moderation API

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/openai-moderation, the October 2026 research run, assessed 1 October 2026. Grade BB, 71.6 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 OpenAI 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. Payments & pricing, 30 out of 100, up to 8.8 more on the total

Why it scored 30: No x402, MPP or L402 (0). The guide says the endpoint is free, and that's public (20). The rate-limits page lists a free usage tier for moderation, but we found nothing saying a new account can call without adding payment details (10 of 20, half for an unconfirmed free start). A person signs up in a browser and makes the key (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.

## 2. Reliability, 65 out of 100, up to 7 more on the total

Why it scored 65: status.openai.com has a Moderations component with 90 days of history (20). Two API-wide incidents in the window list Moderations among the affected components, elevated error rates across API models on 17 September 2026 (1 hour 30 minutes) and elevated errors across ChatGPT, Codex and the API on 29 September 2026 (5 hours 22 minutes, 30 components). The component still reads 100 per cent, but each is an hour or more of wide errors, so two majors (5 of 30, our batch rule for two). Moderation rate limits per usage tier are published, 250 requests a minute on the free tier up to 5,000 on tier 5 (15). The rate-limit and error-code guides cover Retry-After and backoff with jitter (15). The Scale Tier SLA covers GPT and o-series models and doesn't mention moderation (0). omni-moderation 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.

## 3. Maintenance & community, 47 out of 100, up to 4.6 more on the total

Why it scored 47: The newest moderation change is the moderation object on Responses and Chat Completions in the changelog entry of 4 June 2026, about 119 days ago (10). No moderation entries since 3 July (0). Dated changelog several times a month and a help centre, nothing moderation-specific since June (12 of 15). Current official SDKs, openai 3.22.1 on PyPI on 30 September 2026 and openai 7.25.0 on npm (15). SDKs generated from the OpenAPI spec, Python 3.10 to 3.14 (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.

## 4. Agent ergonomics, 85 out of 100, up to 2.4 more on the total

Why it scored 85: A fixed response of flagged, 13 category booleans, 13 scores and the input types each category used, with no field selection (20 of 25). Arrays of inputs in one call, and since 4 June 2026 a moderation object on Responses and Chat Completions returns scores with the generation, but no per-request category choice (10 of 20). The error-code page gives each 401, 403, 429, 500 and 503 a cause and a fix (20). Classification has no side effects and the docs give Retry-After and backoff guidance (20). One required field and official SDKs in Python, JavaScript and other languages (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.

## 5. Security & auth, 92 out of 100, up to 1.4 more on the total

Why it scored 92: Project keys with Restricted and Read-only modes that set None, Read or Write per endpoint, and service-account keys (30). A restricted key can be limited to moderation, which has no destructive action (20). Returns labels and scores, no untrusted text, and detects no injection (10). Usage can be filtered by API key since 4 August 2026, and the organisation has audit logs, but we didn't confirm moderation calls appear in the usage views (12 of 15). security.txt valid, a public bug bounty, SOC 2 Type 2 and ISO 27001 certifications, and the Mixpanel incident disclosed in public (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.

## 6. Schema & documentation, 92 out of 100, up to 1.3 more on the total

Why it scored 92: OpenAPI document in openai/openai-openapi covering /v1/moderations (25). llms.txt at developers.openai.com (10). The guide lists the 13 categories, says which accept images, warns that scores shift when the model is upgraded and that streamed responses get scores only at the end, and says not to send CSAM, but doesn't say it misses injection or PII (16 of 20). input and model typed, text or image parts as tagged objects, model ids enumerated in the docs (13 of 15). Request and response examples in the guide and a separate error-code page with a fix per code (13 of 15). Dated snapshot omni-moderation-2024-09-26 and a dated changelog (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, 90 out of 100, up to 0.9 more on the total

Made of editorial 80, provenance 100.

Why it scored 90: Closed service under the services agreement, SDKs Apache-2.0 (15). The data-controls table lists /v1/moderations as not used for training, no retention by default and eligible for zero data retention, which agrees with the API data policy (30). Deprecations page with dates, including the 27 October 2025 shutdown of text-moderation-007, -stable and -latest (20). Subprocessor list published and data-residency regions listed, but we didn't confirm moderation runs in-region (15 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.

## Deductions

Each comes off the total. A fixed and documented problem counts for less at the next check.

- 2025-11-09, disclosed by OpenAI after notice on 2025-11-25. A breach at Mixpanel, OpenAI's analytics vendor, exposed names, email addresses, coarse location, browser data and organisation and user IDs of platform.openai.com users. No API keys, API requests or usage data were exposed, and OpenAI removed Mixpanel. Fixed and documented, so a small, decayed deduction, the same as other OpenAI API listings in this run (-2). https://openai.com/index/mixpanel-incident/

## 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 a new account can call moderation without adding payment details.
- Whether moderation requests are processed in-region for projects with data residency set.
- Whether the 25 July 2026 API incident also listed Moderations. We didn't read that incident page.

## Weaknesses

- No prompt-injection, jailbreak or PII detection
- One model snapshot from 26 September 2024, and scores can shift when the latest alias moves
- Fixed categories with no custom policies or per-request category choice
- No SLA covers moderation
- Moderations was among the components hit on 17 and 29 September 2026, for about 1.5 and 5.4 hours

## 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.

- Read category_scores rather than flagged alone. The default thresholds are OpenAI's
- Pin omni-moderation-2024-09-26 if the scores feed a decision you audit. The latest alias will move
- Send an array of inputs in one call and match results by index to stay under the per-minute limit
- Add a moderation object to a Responses call instead of a second request when you only need scores on the generation
- Pair it with a separate injection detector. A clean result says nothing about a hidden instruction in a tool result

## What the review panel asked for

- State plainly what it doesn't detect
- an injection category

## 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.