- Conduid
- Marketplace
- #observability
MCP servers tagged observability
791 live MCP servers tagged "observability", ranked by trust score. Tags come from package metadata and repository topics, so this list covers servers across every category that work with observability.
Homelab Evidence MCP
Read-only stdio MCP server for cross-source homelab incident evidence (Gatus, Docker, Loki, Healthchecks).
Drishti
A private, local observability dashboard for Claude Code — see every active skill, MCP server, hook, and agent, your token & dollar usage, context budget, and security findings, l…
Claude Healthline
Never-blank Rust status line for Claude Code that tells you when your agent is going wrong: agent-health score, MCP servers needing attention, context %, live spend.
Spring AI Agents Rag
Production-shaped Spring AI reference app (Java 25, Spring Boot 3.5): multi-provider LLM chat (OpenAI, Vertex AI, Hugging Face, Ollama), advanced RAG over pgvector with Cohere rer…
Zbxmcp
Native Model Context Protocol (MCP) server for Zabbix with an embedded AI assistant. Auto-generates API tools from the installed Zabbix version, enforces RBAC, supports OpenAI/Ant…
Yii3 MCP Telemetry Bridge
Tracing and metrics for MCP servers: spans and RED-style metrics for every yii3-mcp tools/call via rasuvaeff/yii3-telemetry and rasuvaeff/yii3-metrics
ServerProbe
Minecraft 服务器运维探针：启动分析、运行指标、网络取证、代理健康、MCP 深度诊断与 Prometheus，单 jar 支持 Bukkit、Spigot、Paper、Folia、BungeeCord 和 Velocity。
Netdiag Rs
Stateless MCP server for Linux network & system diagnostics over stdio — default-closed gate for privileged probes (rmcp SDK).
Clickhouse Logs
MCP server for ClickHouse: safe read-only SQL queries and cross-service log correlation by session ID and time window
AI API Gateway MCP
Production-style AI risk decisioning gateway with Spring Boot, Spring AI MCP, PostgreSQL, Prometheus, Grafana, and React.
Telemetry MCP
Read-only MCP server querying a configurable metrics backend (BigQuery-ready), schema-agnostic, offline-testable.
Cloud Status Aggregator
Aggregates Azure health from the public status feed, Service Health, and Resource Health into one normalized API, dashboard, and MCP endpoint for AI agents. Deploys to Azure Conta…
Grafana Unified MCP
One MCP server fronting many Grafana instances - adds an 'instance' argument to every upstream mcp-grafana tool
Flutter Lamp
Give your AI live eyes on a running Flutter app — an MCP server over the Dart VM Service: exceptions with stack traces, logs, network, frames, memory, root-cause diagnosis and a l…
Openeuler MCP Toolkit
Read-only openEuler/Linux observability and OS algorithm tools exposed through MCP.
The Skills Barn
326 open, production-ready Agent Skills for the tasks language models can't do reliably from training alone. Free · Apache 2.0.
akkireddy-challa/k8s-mcp-server
inspect pods, deployments, services, and logs via AI agents.
openITCOCKPIT/openITCOCKPIT-MCP-Server
only; the 15 tools that change the monitoring configuration are not registered unless you enable them. Tools take hostnames and template names, not database IDs. Docker image, htt…
Cubiczan/trust-ledger-os
Trust/risk control plane for AI teams: phases, routes, package catalog over MCP. `node packages/mcp/dist/server.js` after build.
Uptime Monitor Worker
Cloudflare Worker: mcp-uptime-monitor. REST + MCP (JSON-RPC 2.0). Live: api.lazy-mac.com/mcp-uptime-monitor
Probesite
Synthetic uptime and Playwright checks exported to Prometheus
Software Observatory
A catalog of epistemic sensors for software correctness — the observable signals that reduce uncertainty about whether a system is correct
First Pass
One agent that catches a film delivery rejection before the platform does.
K8s Observability
Servidor MCP (Model Context Protocol) en Python para inspeccion y telemetria de Kubernetes, Prometheus y Loki mediante agentes de IA.
Aftermath
Assembles a provenance-labelled incident timeline from git, logs and deploy evidence — offline by default, with an MCP server front door.
Activity Mesh
Your second agent session doesn't know what the first one did. Shared activity log for AI agents — auto-captured, auto-injected into context, ~180 tokens, local-first, MIT.
Openmcp Monitor
OpenMCP Monitor — local observability for MCP servers across Claude, OpenCode & the agent stack. Inspect every JSON-RPC frame, track service tokens, manage processes. Zero-depende…
Northstar Workload Optimizer
Governed expense-operations platform with deterministic policy execution, durable workflows, provenance, evaluations, observability, and MCP interfaces.
Tokscope
Token-spend analytics for Claude Code — per-tool, per-task, per-project. CLI + web dashboard + MCP server.
K8s Observability Argocd Server
Safety-first ArgoCD MCP server with progressive disclosure, multi-cluster intelligence, and enterprise-grade observability
Glama AI
Glama — independent third-party profile of a public API surface, by API Evangelist. Glama is an all-in-one AI workspace combining an MCP server marketplace and registry, a managed…
Kafkamcp
Secure, vendor-neutral MCP gateway for operating Apache Kafka with AI agents
Registry Ranking
Private working repository for MCP registry ranking pipeline research, design, and implementation
Cost Meter
Live cost meter for MCP / LLM tool-calling sessions — track token spend per model in real time. Zero runtime dependencies.
Postgres Doctor
Read-only Postgres operational diagnostics over MCP
Observable Agent
A local AI agent that turns self hosted SigNoz into a sensor: it detects an SLO breach through the SigNoz MCP server, applies a policy gated fix, verifies it, and rolls back if it…
OpsMender AI
Open source incident response you run on your own servers. Pages the right people and lets an AI agent investigate and fix things within limits enforced in code.
Tracy
Logs for Ktor services, shaped for a coding agent to read: one Kotlin/Native binary, W3C traceparent correlation and MCP as the primary read path
Kubeshop
Kubeshop — independent third-party profile of a public API surface, by API Evangelist. Kubeshop is the company behind Testkube, an open-core, Kubernetes-native test orchestration…
K5n MCP Hub
MCP gateway with per-user identity: exchanges the caller's token (RFC 8693) so backends see the actual user, not one shared service account. Plus discovery, health checks, tracing…
A MCP Runmeter
MCP server for LLM and agent run telemetry: cost, tokens, latency, and reliability observability
2papi
Self-hosted AI gateway in Go: pool Claude/ChatGPT/Gemini subscription accounts behind one OpenAI-compatible endpoint, with virtual keys, budgets, built-in token savers and an MCP…
ChnAI
AgentWatch — Pingdom for the agent supply chain. Synthetic agent probes continuously test MCP servers for availability, correctness and drift.
Jam
Jam — independent third-party profile of a public API surface, by API Evangelist. Jam is a bug-reporting and debugging-telemetry tool for software teams. Its Chrome extension and…
Enterprise Patterns
Production-grade patterns for enterprise MCP servers — config, structured errors, validation, observability, and a tool registry.
Vpsobs
Read-only observability MCP server: give an agent eyes on a production host, not hands
Neubird
Neubird — independent third-party profile of a public API surface, by API Evangelist. NeuBird AI is "The Production Operations Agent" — an autonomous AI SRE platform that investig…
Glama MCP SDK
Glama MCP SDK for Python/FastAPI: secure client + typed API. TLS, timeouts, retries, pooling. Access 5,000+ MCP servers or run as an MCP server. Optimized for www.glama.ai
Spidey
Production-oriented autonomous coding agent: multi-agent workflows, sandboxed execution, code intelligence, and human approval gates.
Agents Should Survive Failure
Durable, governed AI workflows that survive retries and worker crashes without duplicating consequential business effects.