# Fix list: Payload

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/payload, the October 2026 research run, assessed 8 October 2026. Grade C, 55.2 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 Payload: 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. Security & auth, 57 out of 100, up to 7.5 more on the total

Why it scored 57: REST keys are per user, revocable and regenerable, shown once, and limited by that user's access control down to field level. MCP keys carry per-collection, per-operation switches. Both travel only in the Authorization header. Neither expires, and there is no OAuth (24 of 30). A find-only MCP key is a config switch away and the collection's access control still runs under it. Nothing asks a person to approve a delete or a publish in the free edition (14 of 20). Documents hold text written by other people, and no prompt-injection guidance was found. `overrideResponse` can strip fields before they reach a model (3 of 15). Versions record each change, and `onEvent` and `verboseLogs` expose MCP calls to the owner's own logging. No built-in audit log (7 of 15). SECURITY.md gives a reporting address, and advisories are published on GitHub with affected versions, patches and workarounds. No security.txt, bug bounty or SOC 2 statement was found (9 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.

## 2. Payments & pricing, 45 out of 100, up to 6.9 more on the total

Why it scored 45: Scored with the self-hosted rule. No x402, MPP or L402 (0). The software is free under MIT, which the get-started page states. The only paid option, Enterprise, is priced by sales, and Payload Cloud is closed to new projects, so 15 of 20. No card and no account at Payload are needed (20). `npx create-payload-app` installs it, but the docs create REST and MCP API keys in the admin panel, so half (10).

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.

## 3. Agent ergonomics, 64 out of 100, up to 5.9 more on the total

Why it scored 64: Four MCP tools per enabled collection and two per global, so five collections give 20 tools, the 11 to 30 band (15). We added 7 because the owner enables collections and operations one by one, each key narrows the list again, and find tools default to 10 documents at depth 0 with `select` (22 of 25). Page-based pagination with `limit` and `page`, 17 `where` operators, `sort`, `select` and `populate` (20). REST errors come back in an `errors` array and MCP tools return a text line beginning `Error:`. Neither format is documented as a list an agent can match on (10 of 20). No idempotency keys, and no `readOnlyHint` or `destructiveHint` in the plugin source. An update by id is repeatable (4 of 20). Few required parameters and sensible defaults. The one official client, `@payloadcms/sdk` for TypeScript, is labelled beta (8 of 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.

## 4. Schema & documentation, 70 out of 100, up to 4.9 more on the total

Why it scored 70: No OpenAPI file is published or generated by the core packages, because endpoints depend on each project's config. Every MCP tool carries a typed input schema built from the collection's fields, and the GraphQL schema is generated from the same config (20 of 25). llms.txt per major version, llms-full.txt and a Markdown copy of every docs page (10). Default MCP tool descriptions are one line, such as `Create a document in a collection.`, unless the owner writes one in the plugin config. The docs tell owners to do so (10 of 20). Field inputs are typed with bounds on `limit` and `depth`, but `where` and `select` are JSON passed as strings (10 of 15). REST operations show a request and a response each. No reference page for error responses or status codes was found (8 of 15). Semver, release notes on GitHub and separate docs per major version. REST paths carry no version (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.

## 5. Reliability, 78 out of 100, up to 4.4 more on the total

Why it scored 78: Read with the local-software lines, since the graded surface is software its owner hosts. Official npm packages, with Node.js 18.20.2 or 20.9.0 and later in `engines` and 20.9.0 in the docs (20). A public `ci` workflow runs on every push. On the 3.x branch the 7 October 2026 run passed and the 8 October run failed, and the three latest runs on main failed (17 of 25). 499 issues are open, 140 of them awaiting triage, with 225 opened and 116 closed since 10 July 2026 (12 of 25). A written semver policy marks breaking commits with `!` and lists them in release notes, but lets security fixes and experimental APIs change in minor releases (14 of 15). 3.x is the stable line, with 4.0 in canary (15).

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.

## 6. Transparency & trust, 62 out of 100, up to 3.3 more on the total

Made of editorial 74, provenance 50.

Why it scored 62: MIT for the core and official packages, with LICENSE.md in the repository (30). A self-hosted install keeps content on its owner's servers. The privacy policy dates from 28 March 2024, states no retention periods that we could find and doesn't mention Figma. The only published terms cover Payload Cloud, which is closed to new projects (15 of 30). A written versioning policy defines breaking changes and how they're announced, with a migration guide per major. No end-of-support date for 3.x was found (13 of 20). Telemetry is on by default, documented as anonymous and turned off with `telemetry: false`. The docs don't list the fields sent (16 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: payloadcms.com, registered 2018-04-02 (8 years) (11 of 15)
- Endpoint on the vendor's domain:  is not on payloadcms.com (0 of 15)
- Terms of service: not found (0 of 10)
- Privacy policy: read, states 6 of the 8 things a reader expects (8.5 of 10)
- Status page: not found (0 of 10)
- security.txt: not found (0 of 10)

## 7. Maintenance & community, 78 out of 100, up to 1.9 more on the total

Why it scored 78: 3.90.2 was published to npm on 23 September 2026, 15 days before this check (30). Eight stable releases between 10 July and 23 September 2026, plus 4.0.0 canaries up to 8 October (20). 225 issues opened against 116 closed since 10 July, 140 open issues awaiting triage and 763 open pull requests (12 of 25). `@payloadcms/sdk` is current at 3.90.2 but beta and TypeScript only, and the official MCP registry has no Payload entry, which fits a plugin that runs inside each install (8 of 15). CI runs on every push. The latest 3.x run failed on 8 October 2026 (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.

## Deductions

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

- 18 September to 8 October 2026. Payload published 38 security advisories in three weeks, 49 in the 12 months to 8 October 2026 (8 critical, 23 high, 18 medium). They include remote code execution through the first-register operation (GHSA-97rh-rhh2-7vjv), SQL injection on Postgres and SQLite (GHSA-v49j-62m6-pgrr), unauthorised document updates on orderable collections (GHSA-f7hx-52q9-hcrf) and four in `@payloadcms/plugin-mcp`, one an account takeover through the experimental password recovery tool (GHSA-h5rh-4jwf-738p). All are fixed in 3.90.0 or earlier, each was published by the vendor with affected versions and a workaround, and we found no report of exploitation, so we deduct 10 of a possible 15. https://github.com/payloadcms/payload/security/advisories

## 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.

- The lead omitted the official MCP plugin and Payload Cloud's pause on new projects. Both are recorded.
- No `provenance.terms` is recorded. The only published terms cover Payload Cloud, which is closed to new projects, and the software is under MIT.
- unchecked: a status page. None is linked from the pages we read and status.payloadcms.com didn't resolve.
- unchecked: whether any of the 49 advisories was exploited before its fix. We found no report either way.
- unchecked: Enterprise prices and whether Enterprise adds an audit log. The enterprise page lists what is included without prices.
- unchecked: REST error bodies and status codes by live call. No instance was run.
- LICENSE.md names Payload CMS, LLC, while the terms and privacy policy name Payload CMS, Inc. Figma's legal role isn't stated on the pages we read.
- The versioning policy was read on the main branch (4.x docs). The 3.90.2 tag's docs folder doesn't contain it.

## Weaknesses

- 49 security advisories in the 12 months to 8 October 2026, 8 critical, 38 of them published since 18 September 2026
- No OpenAPI file is published or generated by the core packages, and REST error responses have no reference page
- REST and MCP API keys don't expire, and rate limiting is left to the owner's own hooks or proxy
- No idempotency keys, and the MCP tools set no `readOnlyHint` or `destructiveHint` annotations
- Payload Cloud is closed to new projects, Enterprise is priced by sales, and no status page or security.txt was found

## 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.

- Check the installed version first. Anything below 3.90.0 carries published critical advisories, so ask the owner to upgrade before writing
- Send REST keys as `Authorization: {collection-slug} API-Key {key}` and MCP keys as `Authorization: Bearer {key}`. The two key kinds are separate
- To publish, set `_status: 'published'` in the data. The `draft` parameter only relaxes validation and chooses where an update is written
- Upload files with multipart POST to the upload collection, with other fields as JSON in `_payload`. No MCP upload tool is documented
- Roll back with `POST /api/{collection-slug}/versions/:id` after listing versions. Versions exist only where the collection config enables them

## 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.