Mailtrap Email API + MCP

# Fix list: Mailtrap Email API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/mailtrap, the October 2026 research run, assessed 8 October 2026. Grade B, 66.8 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 Mailtrap Email 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. Payments & pricing, 40 out of 100, up to 7.5 more on the total

Why it scored 40: No x402, MPP or L402 (0). Plan prices and overage per 1,000 emails ($1.00, $0.88, $0.55) published without login, including a Markdown pricing page (20). Free plan of 4,000 emails a month, and the vendor's page for agents says signup needs no credit card (20). A person signs up in a browser and creates the first API token by hand in the UI (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: Graded as a hosted service. Status page at status.mailtrap.info with monthly history by component (20). Since 10 July 2026 it records a sending outage of 1 hour 32 minutes on 1 August covering the Email Sending API, SMTP and Inbound Email, two incidents of about 6 minutes each on 16 September and one of 44 seconds on 27 July on Email Testing, so one major outage (10). Rate limits published with numbers, 150 requests per 10 seconds per token and lower per-account limits on Contacts, Stats and Suppressions (15). The docs ask for exponential backoff on 429 and the specs document `x-ratelimit-reset`, but no idempotency key or safe-retry guidance for sends was found (10). No SLA found in the pricing reference, the terms or the security page (0). The sending API is generally available. The MCP server is at 0.9.0 and Inbound Email has no UI yet (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. Security & auth, 60 out of 100, up to 7 more on the total

Why it scored 60: API tokens with permissions per resource, an expiry date (365 days by default), a reset with a 12-hour overlap and management through the API. The docs give header authentication only (28). Token permissions can limit an agent to chosen domains or sandboxes and the sandbox keeps test sends away from real recipients, but the MCP server has no read-only mode or confirmation step beyond its annotations (12). Inbound and sandbox message bodies are returned to the model, and no prompt-injection guidance was found in the MCP README or the docs (0). Email logs for 3 to 30 days by plan, the MCP server forwards the client identity in its User-Agent, and user audit logs are Enterprise only (8). ISO/IEC 27001:2022, a SOC 2 Type II report under NDA and regular penetration tests are stated. No security.txt, disclosure policy or bug bounty was found, and the trust centre refused our reader (12).

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.

## 4. Agent ergonomics, 62 out of 100, up to 6.2 more on the total

Why it scored 62: 125 MCP tools registered at once, with no toolsets, filtering or dynamic loading (5). List endpoints page by token and per_page or last_id, with search and filters on email logs, sandbox messages and campaigns (17). Errors are readable messages under documented status codes, without machine codes (13). No idempotency key found. All 125 tools are annotated, 53 with readOnlyHint, 43 with destructiveHint true and 29 with destructiveHint false, and none with idempotentHint (12). `DEFAULT_FROM_EMAIL` and `MAILTRAP_SANDBOX_ID` cut required fields, and there are official SDKs in eight languages plus a CLI (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. Schema & documentation, 82 out of 100, up to 2.9 more on the total

Why it scored 82: Graded on the REST API and the MCP server. Ten public OpenAPI 3.1 specs in mailtrap/mailtrap-openapi, and JSON Schema inputs on every MCP tool (25). llms.txt, llms-full.txt and a Markdown copy of each docs page (10). MCP descriptions are mostly one line. A few say when to use the tool or note a quota, and few say when not to (12). Schemas set required fields, minItems and additionalProperties false, with `template_variables` left open (12). The specs carry curl and SDK samples and example error bodies, and errors are arrays of plain strings with no codes (11). Specs are versioned 2.0.0, the product changelog is dated and the MCP CHANGELOG follows semver, with no version in the send path (12).

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. Transparency & trust, 77 out of 100, up to 2 more on the total

Made of editorial 66, provenance 87.

Why it scored 77: Closed service with published terms, and MIT licences on the MCP server, Node SDK and CLI (20). The privacy policy gives retention periods by category (email content and logs within 30 days of a deletion request, account data up to 90 days), and the DPA and sub-processor list agree with it. The policy also allows AI-assisted analysis of email content for abuse prevention (24). No deprecation policy found. The MCP server keeps legacy parameter names, and the terms let Mailtrap modify or discontinue the service with or without notice (4). Twelve sub-processors named with purpose and country, all in the United States, updated 6 October 2026, with data on AWS us-east-1 and Google in the US. EU storage is described as planned (18).

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 6 of the 7 things a reader expects, and has 1 clause that costs points (7.1 of 10)
- security.txt: not found (0 of 10)

## 7. Maintenance & community, 93 out of 100, up to 0.6 more on the total

Why it scored 93: mailtrap-nodejs v5.0.0, the Python SDK 2.11.0 and CLI v0.8.0 were all released on 8 October 2026 (30). MCP 0.5.0 to 0.9.0 between 23 July and 1 September, five Node SDK tags since 14 July and ten dated changelog entries between 28 July and 16 September (20). Pull requests on the MCP repository are merged within days and Dependabot updates land weekly, with 3 open items. A community forum opened on 16 September. Issue reply times weren't visible (18). Listed in the official MCP registry as io.mailtrap/mcp at 0.9.0, and SDKs are current (15). CI runs lint, tests and build on the MCP repository, with CodeQL on the Node SDK (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: trust.mailtrap.io answered 403 to our reader, so the trust centre's list of controls, any disclosure policy there and the SOC 2 request flow were not read
- unchecked: the interactive pricing page at mailtrap.io/pricing is drawn by script. Prices are from the vendor's Markdown pricing reference
- Whether an SLA exists on Enterprise contracts. None is published
- Whether a hosted (remote) MCP server exists. Only the local stdio package was found
- Whether the 1 August 2026 outage has a published post-mortem. The status summary we read gave duration and components only
- The authentication page says an API token has no expiry date, while the API tokens page says new tokens expire after 365 days by default
- Lead check. The lead's interface list was correct, including the Go SDK. The official MCP server is local stdio only

## Weaknesses

- The MCP server registers 125 tools at once, with no toolsets, filtering or read-only mode
- No idempotency key for sends was found in the OpenAPI specs or the docs
- The status page records a sending outage of 1 hour 32 minutes on 1 August 2026 covering the Email Sending API, SMTP and Inbound Email
- Inbound and sandbox message bodies reach the model through the MCP server, and no prompt-injection guidance was found in its README or the docs
- User audit logs are limited to the Enterprise plan, and the free plan keeps email logs for 3 days

## 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.

- Ask the operator to create the first API token at mailtrap.io/settings/api-tokens. The first token can't be created through the API.
- Send transactional mail to send.api.mailtrap.io and bulk mail to bulk.api.mailtrap.io. Sandbox sending uses a different base URL and an inbox ID.
- Verify a sending domain before the first production send, and use Email Sandbox to test payloads without delivery.
- Stay under 150 requests per 10 seconds per token, and 10 per 60 seconds per account on the Stats and Suppressions APIs. On 429, back off until `x-ratelimit-reset`.
- Don't retry a timed-out send blindly, because no idempotency key was found. Check the email logs first.

## 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.