Monitoring · Developer Tools

MCP Coroot

jamesbrink/mcp-coroot

A Model Context Protocol server for Coroot. It lets an MCP client investigate a running system through Coroot: application health, logs, distributed traces, profiles, metrics, SLO incidents, alerts, deployments, costs and configuration.

Install

uvx mcp-coroot

Client configuration

{
  "mcpServers": {
    "coroot": {
      "command": "uvx",
      "args": [
        "mcp-coroot"
      ],
      "env": {
        "COROOT_BASE_URL": "<COROOT_BASE_URL>",
        "COROOT_USERNAME": "<COROOT_USERNAME>",
        "COROOT_PASSWORD": "<COROOT_PASSWORD>"
      }
    }
  }
}

Environment variables

COROOT_BASE_URLCOROOT_USERNAMECOROOT_PASSWORDCOROOT_SESSION_COOKIE
Category
Monitoring, Developer Tools
License
MIT
Updated
Oct 6, 2026

Tools (39)

  • get_connection

    Is Coroot reachable, who are we, and what may we do.

  • save_project

    Create a project or rename one.

  • delete_project

    Delete a project and everything in it.

  • manage_api_key

    Generate or revoke a telemetry ingestion key.

  • list_applications

    Applications with health and failing inspections.

  • get_application

    One application's failing checks and dependencies; name a report for its numbers.

  • get_nodes

    Hosts and their utilisation, or one host's audit report.

  • get_overview

    Project-wide views: map (dependencies), deployments (what changed), risks, costs.

  • get_application_rca

    Coroot's AI root cause analysis, when Coroot Cloud is configured.

  • set_risk_status

    Dismiss a risk as accepted, or restore it.

  • get_logs

    Log entries by severity and text, or repeated messages grouped into patterns.

  • summarize_trace_endpoints

    Per-endpoint request rate, error rate and latency quantiles, worst first.

  • list_traces

    Individual slow or failed traces, with the ids to open them.

  • list_trace_error_reasons

    Top failure reasons with a sample trace id each.

  • explain_trace_latency

    Where the slow tail spends its time.

  • get_trace_by_id

    One trace as a span tree with attributes and events.

  • get_profile

    CPU, memory or lock profile reduced to its hottest frames.

  • get_metrics

    Run a PromQL range query.

  • list_metrics

    Discover metric names.

  • get_incidents

    SLO breaches, or one incident with its burn rates and analysis.

  • get_alerts

    Alerts from Coroot's rules, or one with the charts behind it.

  • set_alert_state

    Resolve, suppress or reopen alerts.

  • get_alerting_rules

    Rules, one rule, or all of them as YAML.

  • save_alerting_rule

    Create a custom rule or replace one.

  • delete_alerting_rule

    Delete a custom rule.

  • get_inspections

    Health checks with their thresholds, or one check's configuration.

  • update_inspection_config

    Change a threshold or define a custom SLO.

  • get_project_config

    Read categories, custom applications, pricing or instance settings.

  • save_application_category

    save_custom_application - Create or update either.

  • delete_project_config

    Remove a category, custom application or integration.

  • get_integrations

    configure_integration - Read or configure notifications and data sources.

  • get_db_instrumentation

    configure_db_instrumentation - Let Coroot collect database statistics.

  • link_telemetry_service

    Map an OpenTelemetry service name onto an application.

  • set_cloud_pricing

    Override the rates used for cost estimates, or reset them.

  • save_dashboard

    Create, rename, or replace a dashboard's panels.

  • delete_dashboard

    Delete a dashboard.

  • get_users

    Users and the roles they can hold, optionally with permissions.

  • save_user

    Create or update a user.

  • delete_user

    Delete a user.

Details on this page are taken from the project's README. Open README

Related MCP servers

More servers
2344 logo

2344

comet-ml/opik

7k

Opik is the open-source LLM observability and evaluation platform for AI agent tracing, LLM evaluation, prompt management, and production monitoring. Built by Comet. Apache-2.0 licensed, free to self-host the full platform, with 20,000+ GitHub stars.

Monitoring
MCP Grafana logo

MCP Grafana

grafana/mcp-grafana

612

A Model Context Protocol (MCP) server for Grafana.

Monitoring
Vite Plugin Vue MCP logo

Vite Plugin Vue MCP

webfansplz/vite-plugin-vue-mcp

407

Vite plugin that enables a MCP server for your Vue app to provide information about the component tree, state, routes, and pinia tree and state.

Monitoring
api200 logo

api200

API-200/api200

169

API 200 is all-in-one gateway for managing third-party APIs efficiently. Integrate third-party APIs or setup MCP server in minutes with auto-generated code, docs, auth, caching and error handling – so you can focus on what really matters.

Monitoring
WireMCP (Wireshark) logo

WireMCP (Wireshark)

0xkoda/wiremcp

98

WireMCP is a Model Context Protocol (MCP) server designed to empower Large Language Models (LLMs) with real-time network traffic analysis capabilities.

Monitoring
gospy logo

gospy

monsterxx03/gospy

90

A tool for inspecting and analyzing running Go processes, including goroutine states, memory statistics, and binary information.

Monitoring