Monitoring MCP servers

A [Model Context Protocol][mcp] (MCP) server for Grafana. This provides access to your Grafana instance and the surrounding ecosystem. Requires uv. Add the following to your MCP client configuration (e.g. Claude Desktop, Cursor): For Grafana Cloud, replace GRAFANA_URL with your instance URL (e.g. https://myinstance.grafana.net). See Usage for more installation options including Docker, binary, and Helm. - Grafana version 9.0 or later is required for full functionality. Some features, particularly datasource-related operations, may not work correctly with earlier versions due to missing API endpoints.

5 tools · Explore agent conversations, get_dashboard_summary

Iris scores every agent run for quality, safety, and cost — on your machine, with no SDK and no account. Most agent projects check quality by running a few remembered prompts and eyeballing the output. Iris replaces that with numbers you can audit: your agent's runs land in a SQLite database on your disk, 13 built-in rules score them deterministically — PII, prompt injection, hallucination markers, cost thresholds — free, with no LLM calls, and an optional LLM judge with a hard per-eval cost cap handles the semantic questions. Every rule is inspectable and editable, because a judge you can't audit is just vibes with a number on it. MIT licensed, no telemetry; your traces never leave your machine.

6 tools · deploy_rule, evaluate_output

Your AI agent doesn't know what's broken in production. This fixes that. Last9 MCP Server connects Claude, Cursor, Windsurf, and any other MCP-capable AI assistant directly to your production observability data — logs, metrics, traces, exceptions, database queries, alerts, and deployments. The agent stops guessing and starts reading the actual signal. No binary to install. No tokens to manage. One URL, OAuth in your browser, done. Find your org slug in your Last9 URL: app.last9.io/<org_slug>/... Type /mcp, select last9, authenticate. That's it.

6 tools · get_alerts, get_database_slow_queries

> [!IMPORTANT] > This repository is archived. The STDIO MCP server in this package is no longer being updated. > We now have a remote MCP server, which allows us to iterate faster on tools and provide a better experience. If you have any questions, reach out to us on Slack or email us at engineering@pydantic.dev.

3 tools · execute_sql, get_spans_by_trace_id

This repository contains th Metoro MCP (Model Context Protocol) Server. This MCP Server allows you to interact with your Kubernetes cluster via the Claude Desktop App! You can read more about the Model Context Protocol here: https://modelcontextprotocol.io But in a nutshell > The Model Context Protocol (MCP) is an open protocol that enables seamless integration between LLM applications and external data sources and tools. Whether you’re building an AI-powered IDE, enhancing a chat interface, or creating custom AI workflows, MCP provides a standardized way to connect LLMs with the context they need.

5 tools · get_cluster_health, get_errors

A remote Model Context Protocol (MCP) server that connects AI assistants to your crash reporting and real user monitoring data in Raygun through natural language conversations. 📚 Tool Reference • 🚀 Quick Start • 🔑 Get API Token - 🐛 Error Management - Investigate, resolve, and track application errors and crashes with full stack traces and context - 🚀 Deployment Tracking - Monitor releases and correlate errors with deployments to identify problematic changes - ⚡ Performance Insights - Analyze page load times, user metrics, and performance trends over time - 👥 User Monitoring - Track customer sessions, behavior patterns, and identify affected users - 🤝 Team Collaboration - Manage invitations and coordinate error resolution across your team - 📊 Metrics & Analytics - Time-series analysis and distribution histograms for errors and performance

6 tools · applications_list, customers_search

Sentry's MCP service is primarily designed for human-in-the-loop coding agents. Our tool selection and priorities are focused on developer workflows and debugging use cases, rather than providing a general-purpose MCP server for all Sentry functionality. This remote MCP server acts as middleware to the upstream Sentry API, optimized for coding assistants like Cursor, Claude Code, and similar development tools. It's based on Cloudflare's work towards remote MCPs. You'll find everything you need to know by visiting the deployed service in production:

5 tools · Investigate errors and issues, Query project and team info

Monitor cron jobs, backups, Kubernetes jobs and other scheduled tasks in Watchgoose.com. Read-only by default, check changes require explicit consent.

AI agent activity logger & monitor MCP server with 20 tools. Post logs, create articles, manage social interactions, and monitor AI agent activities on the Alog platform.

TrustPilot for APIs — independent reliability ratings for APIs and MCP servers, powered by synthetic probes and crowdsourced agent telemetry.

AI agent cost intelligence — track spend across providers, optimize model selection, manage budgets with enforcement, detect cost leaks, and prove ROI. 23 tools across 10 domains.

ThinkNEO Control Plane — Enterprise AI governance MCP server with runtime guardrails, observability, AI FinOps, and agent lifecycle control.

Agents query their own ledger: sessions, costs, loop runs, and stuck-loop flags captured by the Agentic Ledger transparent-proxy flight recorder (local-first, MIT). Install: `pip install agentic-ledger`.

Privacy-first web analytics with native MCP server. Query pageviews, referrers, trends, and AI-generated insights for your sites.

Query, summarize, and trace logs in plain English across GCP Cloud Logging, AWS CloudWatch, Azure Log Analytics, Grafana Loki, and Elasticsearch

Runtime guardrails and incident read access for coding agents. Query AgentGuard traces, alerts, usage, costs, and budget health.

Leverage AI-driven observability, security, and automation to analyze anomalies, logs, traces, events, metrics.

Dead man's switch for AI agents, cron jobs and CI/CD pipelines. An agent creates its own monitor, asks for its ping commands and reports its runs in one conversation; LastPing alerts when a run goes silent, stalls, fails or loops, and `list_open_incidents` hands back what broke with the recurrence count, failing step and exit code. 36 tools. Hosted at `https://mcp.lastping.dev/mcp` (via `mcp-remote`), or `go install github.com/tp322d/lastping-app/cmd/lastping-mcp@latest`. Free for individuals.

Unified agent work logging and observability across ChatGPT, Claude, Cursor, Codex, and OpenClaw with config-first schemas and pluggable sinks.

Two deterministic, rule-based analysis engines for agents: cloud billing cost anomaly detection (rolling-median + MAD, resistant to the masking effect) and auditable architecture risk scoring (single point of failure, coupling, scalability gap) with a rule-by-rule breakdown for every score. `uvx cloudsealed-mcp`

Observability for AI agents: 87 tools to query LLM cost, errors, and latency across OpenAI/Anthropic/Gemini/Mistral, and operate alerts, budget gates, evals, and safety checks from Claude/Cursor. Install: `npx -y @argosvix/mcp-server`.

Query Langfuse traces, debug exceptions, analyze sessions, scores, datasets, schema, observations and manage prompts. Full observability toolkit for LLM applications. (https://github.com/langfuse/langfuse)

eBPF-based GPU causal observability agent with MCP server. Traces CUDA Runtime/Driver APIs and host kernel events to build causal chains explaining GPU latency.

A Prometheus Model Context Protocol Server.

Go-based MCP server for Duplicacy backup monitoring. Query backup job status and Prometheus metrics from a Duplicacy exporter. Docker image available.

AI API cost tracking and budget enforcement across 11 LLM providers. 6 tools for spend analytics, budget monitoring, session summaries, and key management.

Cross-agent failure intelligence. Before an agent retries a failed tool call, `check_tool_failure` reports whether other agents are hitting the same failure right now and which recovery actually worked, with a Wilson-score confidence. Privacy-by-schema: no prompts, keys, or payloads stored. Hosted remote server at `https://failecho.com/mcp`, or self-host.

Gizmo Analytics — cookieless web analytics for AI coding agents. Framework-aware install (Next/Remix/SvelteKit/Nuxt/Astro/Vite/React/HTML), OAuth onboarding, 28 tools for setup + query + workflow operations (summarize_all_sites, explain_traffic_change, detect_anomalies). Hosted MCP at gizmoanalytics.io/mcp.

Read-only Redis monitoring — queue depths by type, Celery queue status, connected clients, server/memory stats, and key counts per database. 5 tools, built with FastMCP. Install: `uvx mcp-redis-monitor`.

Zabbix integration for hosts, items, triggers, templates, problems, data and more.

MCP server providing comprehensive Datadog observability access for AI assistants. Features grep-like log search, APM trace filtering with duration/status/error queries, smart sampling modes for token efficiency, and cross-correlation between logs, traces, and metrics.

VMware Aria Operations monitoring — performance metrics, alarms, capacity analysis, and anomaly detection across vSphere infrastructure. 27 tools (21 read, 6 write) with audit logging for acknowledge/cancel actions.

Official SpeedOf.Me server for accurate internet speed tests via 129 global Fastly edge servers with analytics dashboard and local history

MCP server for Webmin with 61 tools for Linux system administration: services, users, storage, security, databases, and more.

AI-powered observability platform for agents. Auto-discovers and recommends MCP servers; built-in connectors for GA4, Vercel, GitHub, and Sentry with ROI tracking. `npx -y @perceptdot/core`

Local-first LLM API cost tracker. Captures usage across Anthropic, OpenAI, Qwen, and DeepSeek into a local SQLite ledger and exposes spend queries, provider comparison, and recommendations as MCP tools — with first-class Chinese-provider support (CNY→USD). Install: `uvx llm-usage-mcp`.

AI infrastructure cartographer & MCP server: fingerprints, graphs, and watches the live infrastructure an agent can reach (Docker, Kubernetes, cloud, IaC) and detects incidents.

Official remote MCP for RetraceKit. List the project, read issues and incidents, change status, and request AI explanations from Cursor, Claude, Codex, and VS Code.

10 tools · explain_incident, explain_issue

Live .NET runtime diagnostics for AI assistants. Ask Claude to diagnose memory leaks, GC pressure, LOH fragmentation, and thread starvation in any running .NET process — no code changes required. Install: `dotnet tool install -g mcp-dotnet-diagnostics`

Datadog observability — 168 tools across metrics, monitors, logs, APM, RUM, incidents, status pages, fleet automation, workflows. 4 workflow Prompts and `incident-triage-snapshot` aggregation that fans out to 4 fetches in one call.

An MCP server for querying Grafana Loki directly with a discovery-first workflow — labels, values, series, and LogQL queries without requiring Grafana.

HTTP uptime monitoring, heartbeat (dead-man's switch) monitoring, cron job scheduling, and email verification. Single API key, per-tool quota tracking. Install: `go install github.com/tickstem/mcp/cmd/tsk-mcp@latest`.

Unofficial FortiMonitor v2 API integration with 241 tools for server monitoring, outages, maintenance, metrics, notifications, and more.

Check real-time operational status of 114+ cloud services and APIs (AWS, GitHub, Stripe, OpenAI, Vercel, etc.) directly from AI assistants. Published on npm.

A system monitoring tool that exposes system metrics via the Model Context Protocol (MCP). This tool allows LLMs to retrieve real-time system information through an MCP-compatible interface.（support CPU、Memory、Disk、Network、Host、Process）

Data freshness verification for AI agents. Probes endpoints for HTTP cache staleness, latency percentiles, content fingerprinting, TLS health, DNS timing, and redirect chains. Returns deterministic FRESH/STALE/UNKNOWN verdicts with NIST AI RMF mapping. CLI + MCP server + HTTP API with Prometheus metrics and YAML policy engine.

Read-only log access for AI assistants over HTTPS. Whitelist log files on your Linux server; AI can search and read them without shell access. Token-authenticated, syslog-audited.

Measures whether a Claude subscription is paying off, from the logs Claude Code already writes on disk: how much of the code Claude wrote is still in your files, what actually drives the rate limit, and how much of that limit plain chat ate. Single Python file, no dependencies, nothing leaves the machine.

Server ops via AI: log troubleshooting, CPU/memory/disk monitoring, code edit, Nginx & certificate management. Local or SSH-remote with two-step confirmation, command whitelist, and secret redaction. Install: `npx -y server-ops-mcp`.

Agent tool spend tracker. Local JSONL storage, consent-gated, nothing leaves the file. Free MCP server. `npx scvd-tab`

TypeScript MCP server for the Langfuse Public API. 27 read tools covering traces, observations, sessions, scores, score-configs, prompts (with version/label), datasets, dataset items, dataset runs, metrics, models, projects, comments, media, and health. Distributed as `npx -y langfuse-mcp` with provenance-signed releases.

An MCP server that allows querying Loki logs through the Grafana API.

Internet speed testing with network performance metrics including download/upload speed, latency, jitter analysis, and CDN server detection with geographic mapping

Fast and minimal Sentry MCP server written in Rust

Provides comprehensive integration with your [VictoriaMetrics instance APIs](https://docs.victoriametrics.com/victoriametrics/url-examples/) and [documentation](https://docs.victoriametrics.com/) for monitoring, observability, and debugging tasks related to your VictoriaMetrics instances

Model Context Protocol server for monitoring Operational Status of major digital platforms in Claude Desktop.

Read-only MCP server for OpenObserve Community Edition via REST API. Search logs, traces, stream schemas, and dashboards without requiring the Enterprise license.

AI-native web analytics. Query pageviews, top pages, referrers, countries, devices, and custom events. Create conversion funnels and alerts.

Query Spanlens LLM observability from Claude Desktop, Cursor, or Continue: request logs, costs, agent traces, anomalies, and user analytics with a read-only API key.

Local system monitoring — CPU, memory, swap, disk, network, and process management. Find resource-hungry processes, diagnose performance issues, and kill processes via AI.

Transparent proxy for any MCP server that intercepts all JSON-RPC messages, measures latency, and stores traces in SQLite, PostgreSQL, or MySQL. Exports OpenTelemetry (OTLP) spans to Jaeger, Tempo, or Grafana. Supports stdio, HTTP, and SSE transports. `npx @cardor/heimdall-mcp`

Prices every Claude Code and Codex session already on disk and lets the agent ask what it's spending mid-session: cost by session, sub-agent and model, week-on-week comparison, and plain-English findings with applyable patches. No proxy, no API key, nothing leaves the machine. Install: `brew install --cask magna-nz/tap/tallybook`.

Debug your Container and Kubernetes workloads with an AI interface powered by eBPF.

Go-based MCP server that bridges any GenieACS (TR-069 ACS) instance, exposing device data, firmware management, and CPE actions (reboot, parameter refresh, firmware download) over JSON-RPC. Docker image available.

Instant website audits with 11 tools — full SEO audit (20+ checks), security headers and SSL verification, Lighthouse performance metrics, multi-site comparison, broken link checker, WCAG accessibility audit, Schema.org structured data validation, competitor gap analysis, and robots.txt parsing. No API keys required.

Full-coverage MCP server for Umami Analytics API v2 — 66 tools for websites, stats, sessions, events, reports, users, teams, and realtime monitoring.

Tautulli (Plex media server monitoring) with 11 read-only tools for activity, history, library stats, user stats, transcode analysis, and resolution breakdowns.

Discovery, exploration, reporting and root cause analysis using all observability data, including metrics, logs, systems, containers, processes, and network connections

Comprehensive MCP server for Grafana, Prometheus, Kafka UI, and Datadog with a secure "Bring Your Own Key" or BYOK model.

Synthetic monitoring for LLM inference endpoints. Measure TTFT, latency, throughput, and errors across OpenAI, Anthropic, Google, Azure, Bedrock, and local servers (vLLM, SGLang, Ollama). CLI + MCP server with Prometheus and OpenTelemetry export.

All-in-one homelab management MCP server. Monitor system resources, manage Docker containers, Wake-on-LAN, scan networks, check open ports, and run alerts — across multiple servers via SSH. Single 10MB binary, zero dependencies.

🎖️ 🏎️ ☁️ – Interact with Edge Delta anomalies, query logs / patterns / events, and pinpoint root causes and optimize your pipelines.

Read-only MCP server for on-prem Linux VMs and PostgreSQL over SSH. Check service health (systemd/journald), retrieve bounded logs, inspect DB state, and explore table schemas — without terminal access. Allowlist-gated, append-only audit log. `pip install infra-mcp`.

Socket.IO-style telemetry for MCP servers. Instrument a tool call with a few lines of `mcp-telemetry-sdk`, and `telemetry_subscribe` streams its live progress (steps, logs, cost, done) to any connected MCP client via `notifications/progress` — no polling, and a job started in one session can be watched from a completely different one.

Tamper-evident observability for AI agents: a SHA-256 hash-chained audit log with chain verification and signed export (EU AI Act Art. 12). Instrument any agent with zero code via `npx -y @agentlensai/mcp`; also ingests OpenTelemetry GenAI traces.

Zero-config data quality monitoring across Postgres, BigQuery, Snowflake, MySQL, and DuckDB. Profile a warehouse, detect anomalies (volume, schema drift, freshness, NULLs, distribution, cardinality), with optional dbt manifest lineage. Read-only — connection resolved server-side, never passed via the model.

MCP server that checks the live status of 3831 software services in real time. Ask your AI agent "is GitHub down?" or "what's wrong with Sentry?" — and get a live answer pulled directly from official status pages, including full incident detail when something is broken. `npx -y github:jabbawocky/statuscraft`

Governed managed-endpoint fleet operations (thin-client/VDI) — login-storm and patch/config-drift analysis, inventory, and guardrailed remediation (13 tools) with unbypassable audit logging (MCP + CLI), budget/runaway guards, and undo/rollback.

Query Langfuse traces, debug exceptions, analyze sessions, and manage prompts. Full observability toolkit for LLM applications.

Real-time cost observability for MCP tool calls. Transparent proxy auto-detects every MCP client (Claude Desktop, Cursor, Windsurf, VS Code, Claude Code, Zed, Continue.dev, Cline, Goose) and attributes spend per tool, per project, per end-customer. `npx @mcpspend/proxy add` install. Free tier 25K calls/month, no card. MIT proxy on npm. EU-hosted, GDPR-ready.

Enhance AI-generated code quality through intelligent, prompt-based analysis across 10 critical dimensions from complexity to security vulnerabilities

Run efficiency for Claude Code sessions: the share of paid tokens that became model output versus re-reading context already sent, by tokens and cost-weighted. Reads local transcripts, makes no network calls.

Website uptime monitoring from 300+ global locations. Run instant HTTP, ping, port, DNS, WHOIS, blacklist, Web Risk and page-speed checks, and manage monitors, contacts, webhooks, incidents, maintenance windows and status pages. Remote streamable-HTTP server at `https://mcp.host-tracker.com/mcp`, bearer token.