Servidor MCP de Zenofex LLC

• Nombre en el registrors.monit/monit-rsverificado en el registro MCP

El nombre en el registro está bajo monit.rs. El registro solo permite usar ese espacio de nombres a quien demuestra control del dominio por DNS o HTTP.

• Estadorespondecomprobado el 2026-10-20
• Herramientas22
• El servidor se llamamonit.rs
• Versión del servidor1.28.1
• Versión del protocolo2025-06-18

El 2026-10-20 el servidor respondió a initialize y tools/list sin iniciar sesión. Primera respuesta en 52 ms.

• Aún no se ha juzgado si sus herramientas hacen el trabajo del producto.

Lo dicen reglas a partir de las mediciones de arriba. La primera llamada probable la elige un modelo de la lista de herramientas; lee su descripción antes de fiarte.

• Nota INITE MCP73 de 100responde 100 · definiciones 70 · contexto 100 · conexión 100 · rebajada por las señales de aviso de abajo
• Respondióel 100% de los días14 de 14 comprobaciones diarias
• Mediana de la primera respuesta79 ms
• Espacio en el contexto del agente≈ 2602 tokenslas definiciones de todas sus herramientas (22), que el agente lleva en cada turno
• Definiciones de herramientasC · 70 de 100
• Resultados tipados9 de 22 herramientasdeclaran outputSchema: el resultado llega tipado y se lee de una vez

• Parámetros sin descripción: en 17 de 22
• Sin indicar si solo lee o puede destruir: 22 de 22

Un programa comprueba el servidor una vez al día: initialize y tools/list, sin modelo. Los tokens son la longitud de las definiciones, a unos cuatro caracteres por token. La nota promedia seis comprobaciones del nombre, la descripción, el esquema y las anotaciones de cada herramienta; dice lo bien descritas que están, no si funcionan. La nota pondera solo lo medido: si responde (25%), si hace el trabajo del producto (25%), definiciones (20%), contexto (10%), uso (10%) y lo fácil que es conectarse (10%). Lo que no se midió queda fuera, y una instrucción oculta limita la nota a 40. El estado de los servidores MCP de los proveedores: método y números →

• openapi_diff

  Compare two OpenAPI 3.x specifications and classify every change as breaking, non-breaking, or info. Use this when you want to know if a candidate OpenAPI spec change is safe to ship or will break existing clients. The diff covers paths…

• monit_rs_stats

  Get live aggregate statistics for the monit.rs platform. Returns approximate counts of regressions caught, AI analyses generated, and probes executed across all monitored APIs to date. Useful when answering "does monit.rs actually work at…

• monit_rs_describe

  Return the canonical product description for monit.rs. Use this when a user asks what monit.rs is, what it does, how it differs from Pingdom / Checkly / UptimeRobot, or for product URLs. The response is the curated content from…

• list_my_endpoints

  List all monitored endpoints owned by the authenticated account. Returns brief records suitable for LLM enumeration. For full detail (headers, spec URL, heartbeat_token) use `describe_endpoint(id)`.

• list_my_incidents

  List incidents on the caller's account. Defaults to only OPEN incidents. Fields include the incident's dedup_key — pass this to `acknowledge_incident` or `resolve_incident` to act on it.

• list_my_regressions

  List detected schema regressions on the caller's account. Optionally filter by endpoint_id.

• get_endpoint_uptime

  Aggregated uptime summary for an endpoint over the requested window.

• describe_endpoint

  Full detail for one endpoint. Use `list_my_endpoints()` to get the ID.

• acknowledge_incident

  Mark an incident as acknowledged. Use `list_my_incidents()` to find the dedup_key.

• resolve_incident

  Close an incident. Symmetric to acknowledge_incident.

• create_endpoint

  Create a new monitored endpoint. Respects your tier's max_endpoints cap.

• set_endpoint_assertions

  Replace an endpoint's body assertions. Each assertion is {"path": "$.ok", "op": "equals", "value": true}. Operators: equals, not_equals, exists, not_exists, gt, gte, lt, lte, contains, not_contains, empty, not_empty. A failing assertion…

• delete_endpoint

  Delete a monitored endpoint. Cascades: removes test_results, baselines, regressions, incidents attached to it. Guarded by two safety layers beyond the scope check: 1. `confirm_by_typing_the_endpoint_name` MUST equal the endpoint's exact…

• list_alert_channels

  List alert channels configured on the authenticated account. Does NOT return decrypted config (webhook URLs, PagerDuty routing keys, etc.). Use the dashboard or `describe_alert_channel` HTTP API when you need to inspect the actual config.

• list_status_pages

  List public status pages owned by the authenticated account.

• list_security_findings

  List security findings detected on the caller's monitored endpoints. Optionally filter by severity.

• describe_incident

  Full detail for one incident. Use `list_my_incidents()` to find the dedup_key. Includes the incident's endpoint name and current status.

• pause_endpoint

  Pause a monitored endpoint (stops probing, keeps history). Idempotent — no-op if already paused.

• resume_endpoint

  Resume a paused endpoint. Rejects if resuming would exceed your tier's active-endpoint cap (upgrade or pause another endpoint first).

• update_endpoint_interval

  Change how often an endpoint is probed. Minimum interval varies by tier (free=900s, developer=300s, starter=60s, pro/scale/enterprise=30s).

• create_status_page

  Create a public status page at https://<slug>.status.monit.rs. Add endpoints to it via the dashboard or the HTTP API (adding endpoints from chat is deliberately not exposed — the mapping is fiddly and easy to get wrong).

• create_alert_channel

  Create an alert channel. Pass EXACTLY one of email_address / webhook_url / pagerduty_routing_key matching the channel_type: - channel_type='email' → email_address='[email protected]' - channel_type='webhook' →…