MCP server for Samotpravil SMTP API — documentation search and API proxy. Interim npm: samotpravil-mcp; future: @samotpravil/mcp after org migration.
46 tools with mixed quality. Naming is consistent (verb_noun pattern: get_, list_, send_, search_, create_, update_, delete_). Descriptions present but brief (avg ~40-80 chars, well below 194-char baseline). Input schemas visible and typed, but many parameters lack descriptions or have minimal ones (e.g., 'Dry run mode' for dry_run across 5+ tools). Output schemas not documented. Error handling absent, no recovery guidance, no actionable error messages. Tool composition is reasonable (single responsibility), but parameter descriptions are sparse. STDIO transport caps protocol readiness at 50.
Добавить email в стоп-лист
Generic HTTP к api.samotpravil.ru (safety flags, dry_run). Отключается через SAMOTPRAVIL_ALLOW_GENERIC_API=0.
POST /api/v2/authkey/create — WhiteLabel (Python SDK parity)
POST /api/v2/blist/create — WhiteLabel (Python SDK parity)
POST /api/v2/blist/domains/add (Python SDK parity)
POST /api/v2/blist/domains/verify (Python SDK parity)
Parameter descriptions are minimal or missing. Most parameters have 1-3 word descriptions (e.g., 'Dry run mode', 'Email to search'). Baseline is 72 chars; these average ~20-30. LLMs cannot infer parameter semantics from names alone.
No output schemas documented. Tools return data but LLMs cannot plan downstream calls or extract fields without seeing response structure. Baseline: 100% of A+ tools document return types.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 58 | 2026-07-28+ | v2 |
POST /api/v2/blist/domains/remove (Python SDK parity)
GET /api/v2/authkey — WhiteLabel (Python SDK parity)
GET /api/v2/blist — WhiteLabel (Python SDK parity)
GET /api/v2/issue/status по x_track_id
Документация по методу API.
GET /api/v2/issue/ext_status — полная история статусов (Python SDK parity)
GET /api/v2/blist/report/fbl (Python SDK parity)
GET /api/v2/issue/report/fbl (Python SDK parity)
GET /api/v2/ip — WhiteLabel (Python SDK parity)
GET /api/v1/get_issue_stat (Python SDK parity)
GET /api/v2/blist/report/non-delivery (Python SDK parity)
GET /api/v2/issue/report/non-delivery (Python SDK parity)
Обзор API СамОтправил.
GET /api/v2/package/status
GET /api/v2/issue/statistics (Python SDK parity)
GET /api/v2/blist/report/unsubscribe (Python SDK parity)
GET /api/v2/issue/report/unsubscribe (Python SDK parity)
GET /api/v2/blist/domains
Список методов API.
Diff Postman API vs локальный data/collection.snapshot.json.
Коллекция документации из Postman API (summary или full JSON).
Поиск запросов в коллекции (local snapshot или remote Postman API).
Postman API → data/collection.snapshot.json (write=true для записи).
Удалить email из стоп-листа
Поиск по документации API.
Поиск email в стоп-листах
POST /api/v1/smtp_send
POST /api/v2/mail/send
POST /api/v1/add_json_package — массовая отправка (Python SDK parity)
GET /api/v1/add_package — XML-пакет по URL (Python SDK parity)
POST /api/v2/stop-list/export — создать задачу экспорта (Python SDK parity)
DELETE /api/v2/stop-list/export/{token} (Python SDK parity)
GET /api/v2/stop-list/export/{token} (Python SDK parity)
GET /api/v2/stop-list/export — список задач экспорта (Python SDK parity)
GET /api/v2/stop-list/failed (Python SDK parity)
GET /api/v2/stop-list/fbl (Python SDK parity)
GET /api/v2/stop-list/unsubscribe (Python SDK parity)
GET /api/v1/package_stop — остановить пакет (Python SDK parity)
POST /api/v2/blist/update — WhiteLabel (Python SDK parity)
POST /api/v2/emails/validate/
No error handling or recovery guidance. Tools lack descriptions of failure modes, retryability, or actionable error messages. Agents cannot self-correct on failures.
Destructive tools (send_email, send_package, stop_package, domain_remove) lack dry_run or confirmation patterns. Agents can cause irreversible side effects without safeguards.
Tool descriptions are too brief (avg 40-80 chars vs 194-char baseline). Many lack context on WHEN to use the tool vs similar ones (e.g., send_email vs send_mail_v2 distinction unclear).