# Fix list: Mailchimp Marketing API From Anchor Terminal's listing at https://www.anchorterminal.com/tools/mailchimp-marketing, the October 2026 research run, assessed 8 October 2026. Grade C, 57.4 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 Mailchimp Marketing 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. Security & auth, 46 out of 100, up to 9.5 more on the total Why it scored 46: API keys are named, several to an account, revocable, and bounded by the creating user's level. A key has full account access, and the docs say there is no read-only key and no per-endpoint limit. Keys made from 22 June 2026 expire after one year per the help centre. OAuth 2 uses the authorisation code grant with no scopes, and its tokens never expire. Credentials travel in a header, not the URL (20). No read-only mode. `send-checklist`, `actions/test`, `actions/unschedule` and `actions/cancel-send` exist, with no approval step before a send or a permanent contact delete (6). The API returns contact fields, conversation messages and survey answers, which are untrusted text, and no guidance on prompt injection was found (2). `/authorized-apps` lists the OAuth apps on an account and each response carries a request id. No API call log or audit log was found in the docs we read (4). A responsible disclosure programme with no rewards, a security.txt with a contact and no Expires field, SOC 2 reports, ISO 27001, regular penetration tests and optional two-factor sign-in. No public advisory feed found (14). 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, 25 out of 100, up to 9.4 more on the total Why it scored 25: No x402, MPP or L402 (0). Plan prices are public without a login, from $13 a month for Essentials, $20 for Standard and $350 for Premium with USD selected, rising with contacts through a selector. Nothing is priced per API call (10). The Free plan has 250 contacts and 500 sends a month with no time limit. We could not confirm that signup asks for no card, and the 14-day trial of a paid plan needs payment details, so 15 of 20. A person signs up in a browser and creates a key or approves OAuth (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. ## 3. Reliability, 60 out of 100, up to 8 more on the total Why it scored 60: Graded on the hosted REST API with the hosted lines. status.mailchimp.com lists an API component for each of 21 data centres and keeps event history by month (20). In the 90 days to 8 October 2026 it shows five events on the marketing side. Click tracking failed from about 11 am Eastern on 3 August to noon on 4 August and cannot be backfilled, campaign links returned 500 errors on 13 August, campaign reports showed 0 per cent on 4 August, contacts on some accounts did not enter flows on 22 July, and audience updates were delayed for about two and a half hours on 11 September. None names the API component. We count the day of lost click data as one major (10). The docs give 10 simultaneous connections per user, a 120-second timeout and 500 pending batches, with no per-minute figure (12). 429 is documented with its cause. No `Retry-After` header, backoff guidance or idempotency key was found in the docs, though the new client libraries retry 408, 429 and 5xx twice with exponential backoff and PUT on a member is an upsert (8). No SLA found (0). Version 3.0 is generally available, and the Audiences endpoints are beta (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. ## 4. Agent ergonomics, 73 out of 100, up to 4.4 more on the total Why it scored 73: Graded on the API, since the MCP server only searches docs. `fields` and `exclude_fields` trim responses on 130 operations (20). `count` up to 1,000 with `offset` on 59 operations and per-endpoint filters. No cursor, so deep pages get slow against the 120-second timeout (16). Errors are `application/problem+json` with `title`, `status`, `detail`, `instance` and a per-field `errors` list, and `X-Trigger-Error` returns a named error on demand (17). No idempotency key. PUT on a member upserts by email hash, `is_syncing` keeps loaded events from triggering automations, and scheduled campaigns can be unscheduled (8). One root URL and few required fields, but a contact is addressed by an MD5 hash of the lowercase email. Nine official client libraries and a CLI, with the TypeScript package missing from npm and PyPI still on the 2022 library on the day (12). 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. Schema & documentation, 76 out of 100, up to 3.9 more on the total Why it scored 76: An OpenAPI 3.1.1 file with all 297 operations is public on the API host, and the docs site serves a 1.2 MB copy (25). llms.txt at each level of the docs and a Markdown copy of every page at the URL plus `.md` (10). Every operation has a description, averaging about 60 characters. They say what a call does and seldom when not to use it, and four agent skills cover consent, OAuth, webhooks and bulk jobs (10). The smaller file carries 498 enums and required fields on request bodies. Merge fields and event properties are free-form objects (12). The API's own file has examples throughout and a `default` error response on every operation, and the docs have an error glossary. The docs-site copy lists only 200 and 204 responses (12). The version is in the path and has been 3.0 since 2016. The public changelog's newest entry is 19 August 2025, so key expiry from 22 June 2026, webhook signing and the new client libraries are not in it, and the specification's version label is 0.1.0-dev (7). 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. Maintenance & community, 63 out of 100, up to 3.2 more on the total Why it scored 63: Release 1.0.3 of the TypeScript, Python and Go client libraries is dated 5 October 2026, three days before this check (30). Dated releases in the last 90 days are the client libraries on 5 October, the CLI's 0.0.212 and 0.0.213 on 18 September, Ruby 0.0.218 on 18 September and Rust 0.0.221 on 21 September, all from one generator launch that began on 9 September. The API changelog has no entry in that time, so 15 of 20. A public changelog whose newest entry is 19 August 2025, an API help address, and a help centre article saying support is not trained for in-depth API questions. We saw no public issue replies (5 of 15). Nine official client libraries generated from the current specification, with TypeScript not on npm and PyPI at 3.0.80 from November 2022 on the day (8 of 15). Each repository has a workflow that compiles and tests, whose results we could not see, and release 1.0.3 of the TypeScript library removed the `LICENSE` and `NOTICE` files added on 18 September (5). 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. ## 7. Transparency & trust, 74 out of 100, up to 2.3 more on the total Made of editorial 55, provenance 92. Why it scored 74: Closed service with published terms, an API Use Policy and a named contracting entity. The client libraries declare Apache-2.0 in package metadata (15). A data processing addendum effective 26 September 2024 covers deletion or return on termination without a stated period, and the Intuit Global Privacy Statement of 9 March 2026 is the privacy document. The terms let Mailchimp use inputs and outputs of its AI functions, customer data included, for machine learning, with a preference in account settings. No retention periods for marketing data were found (18). No deprecation policy was found. The API Use Policy says the APIs can change by a posting or an email. Dated notices exist, such as key expiry from 22 June 2026 in the help centre (8). A sub-processor list with locations is public and was last updated on 29 February 2024, and the security page places Mailchimp's own servers in United States data centres (14). 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): - Terms of service: read, states 5 of the 7 things a reader expects, and has 2 clauses that cost points (4.3 of 10) - Privacy policy: read, states 8 of the 8 things a reader expects, and has 1 clause that costs points (8 of 10) ## Deductions Each comes off the total. A fixed and documented problem counts for less at the next check. - 8 October 2026. The quick start tells users to run `npm install @mailchimp/mailchimp-marketing`, and the npm registry answers 404 for that name. Its `pip install mailchimp-marketing` installs version 3.0.80 from November 2022, which has no `MailchimpClient` class as the quick start's code needs. The Go, Ruby, Rust and PHP names resolve, so 2 for a claim the listing can't meet in two of nine languages (https://mailchimp.com/developer/marketing/build/start-developing/send-your-first-api-request.md, https://registry.npmjs.org/@mailchimp%2fmailchimp-marketing). ## 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: whether signing up for the Free plan asks for a payment card - unchecked: paid prices above the starting contact tier, which the pricing page draws by script from a contact selector. The $13, $20 and $350 starting prices are the page's values with USD selected - unchecked: which API functions each plan allows. The docs say access depends on the plan without a table - unchecked: any API call log or audit log inside the account, which is behind a login - unchecked: GitHub stars, open issues and workflow results for the client library repositories, which were read by shallow clone only - unchecked: whether the Java and C# packages named in the quick start are on Maven Central and NuGet. npm, PyPI, crates.io, RubyGems and Packagist were checked - unchecked: the detail pages of the status events, so scope and cause beyond the summaries are not known - unchecked: SOC 2 and ISO reports, which sit in the Intuit compliance portal - Security incidents in the last 12 months were not searched for beyond the vendor's own pages, because web search was not available - The developer docs say API keys don't expire. The help centre says keys generated from 22 June 2026 expire after one year. We took the help centre's dated statement as current - The llms.txt header gives the MCP server as https://mailchimp.com/_mcp/server and the docs pages give https://mailchimp.com/developer/_mcp/server. Only the second was tested, and it answered - The official MCP registry has no Mailchimp entry under a vendor namespace. Third-party servers use the name - The privacy field points at Intuit's Global Privacy Statement because Mailchimp publishes no privacy policy of its own, as on the Mailchimp Transactional listing - Docs pages carry lines addressed to AI clients that point at llms.txt and the MCP server. Recorded as a fact, with no deduction ## Weaknesses - An API key has full account access. The docs say there is no read-only key and no way to limit a key to chosen endpoints - OAuth 2 tokens carry no scopes and never expire, and no approval step before `actions/send` was found - The quick start's `npm install @mailchimp/mailchimp-marketing` answered 404 on npm, and `pip install mailchimp-marketing` installs a 2022 library with a different client class - The only stated limit is 10 simultaneous connections per user. No `Retry-After` header, idempotency key or SLA was found - The developer changelog's newest entry is 19 August 2025, and the docs say keys never expire while the help centre says new keys expire after one year - Click tracking was lost for about 25 hours on 3 and 4 August 2026 and cannot be backfilled, per the status page ## 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. - Send the key or OAuth token as `Authorization: Bearer` to `https://api.mailchimp.com/3.0/`. Treat it as full account access and keep it server-side - Address a contact by the MD5 hash of the lowercase email address, and use PUT on `/lists/{list_id}/members/{subscriber_hash}` so a retry does not create a duplicate - Ask a person before POST `/campaigns/{campaign_id}/actions/send`. Read `/campaigns/{campaign_id}/send-checklist` and use `actions/test` first - Keep to 10 concurrent requests per user, and send large jobs to `/batches`. A 429 names no wait time, so back off and retry - Add `fields=` to every list call and set `count` (default 10, maximum 1,000) with `offset` - Set `is_syncing` to true when loading historical events so they do not trigger automations ## 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.