Read-only MCP server for the Lettermint email platform (Team API)
Lettermint MCP demonstrates solid definition quality with consistent naming conventions, comprehensive parameter schemas using Zod, and clear descriptions across all 20 tools. All tools follow verb_noun naming (get_, list_) and include descriptions (avg ~120 chars). Zod schemas provide type safety and validation. However, output schemas are not explicitly documented in tool definitions, LLMs cannot see what fields to expect from responses. Error handling is basic (LettermintError caught and surfaced as tool errors) but lacks recovery guidance. No destructive operations present (all READ_ONLY), reducing risk. Pagination is well-implemented across list tools with cursor-based navigation.
Get details of a domain, optionally including its DNS records (SPF/DKIM/etc with verification status) and associated projects.
Get full details of a message: sender, recipients, subject, status, attachments, metadata, spam score and spam symbols.
Get the content of a message: raw MIME source, rendered HTML body, or plain text body.
Get the event timeline of a message (queued, delivered, opened, clicked, bounced, ...) with timestamps and metadata.
Get details of a project, optionally including its routes, domains, and team members.
Get details of a route: type, settings, inbound configuration, and optionally its project and statistics.
Output schemas not documented in tool definitions. LLMs cannot see what fields responses contain, forcing them to guess at downstream field names and risk broken tool chains.
Error messages lack recovery guidance. LettermintError is caught and surfaced as tool errors, but messages do not suggest next steps (e.g., 'Try search_domains() with a partial name' if domain not found).
Parameter descriptions for include arrays lack clarity on what each option returns. E.g., get_domain's include=['dnsRecords','projects'] does not explain what fields each adds to the response.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 61 | 2026-07-28+ | v2 |
Get aggregated message statistics (sent, delivered, bounced, opened, clicked, spam complaints, inbound) with daily breakdown for a date range of at most 90 days.
Get details about the Lettermint team: name, plan, tier, verification status, and counts of domains/projects/members.
Get current and historical email usage with billing periods (up to 12 most recent periods).
Get details of a webhook: URL, subscribed events, enabled status, last call time. The signing secret is redacted unless include_secret is true.
Get full details of a single webhook delivery attempt, including request payload, response body/headers, and error message.
List sending domains for the team with their verification status.
List messages (sent and inbound emails) for the team, with rich filtering. Use this to inspect email activity, find bounces, or debug delivery issues.
List projects for the team with route/domain/member counts and last-28-days statistics.
List routes of a project. Routes define how emails are processed: transactional, broadcast, or inbound.
List suppressed recipients (addresses Lettermint will not send to) with the suppression reason and scope.
List team members with their user info, role, and join date.
List delivery attempts of a webhook with status, HTTP status code, attempt number, and duration. Use this to debug failing webhooks.
List webhooks configured for the team, with their subscribed events and enabled status.
Ping the Lettermint API to verify connectivity and that the token is valid.
Date format constraints in descriptions are informal. 'Y-m-d format, e.g. 2026-06-01' relies on example parsing; should use JSON Schema format:'date' or explicit regex pattern.
No tool composition guidance. Descriptions do not hint at common multi-step workflows (e.g., 'Call list_messages() first to find a message_id, then get_message_events() to see delivery timeline').