{"slug":"thotischner/observability-mcp","name":"observability-mcp","description":"One MCP server that connects to any observability backend through pluggable connectors, normalizes the data, adds intelligent analysis, and provides a web UI for configuration.","category":"other","tags":[],"official":false,"stars":4,"transport":"stdio","install":[{"cmd":"npx @thotischner/observability-mcp","imports":[]}],"tools":[{"name":"list_sources","description":"Discover configured backends and connection status"},{"name":"list_services","description":"Discover monitored services across all backends"},{"name":"query_metrics","description":"Query metrics with pre-computed summary stats"},{"name":"query_logs","description":"Query logs with error/warning counts and top patterns"},{"name":"get_service_health","description":"Health score combining metrics + logs (0–100)"},{"name":"detect_anomalies","description":"Cross-signal anomaly detection with robust (median/MAD + trend) analysis"},{"name":"get_topology","description":"Return the merged infrastructure graph (resources + edges) from every topology-capable connector, filterable by source/kind/scope"},{"name":"get_blast_radius","description":"Pivot on the universal RUNS_ON relation — 'if this resource's host fails, who else fails?'. Works for pod→node, vm→hypervisor, container→host"}],"env_vars":["PROMETHEUS_URL","LOKI_URL","GRAFANA_TOKEN"],"auth_type":"none","github":"https://github.com/ThoTischner/observability-mcp","homepage":"","server_url":"","status":"active","source":"mcpservers.org","updated_at":"Thu May 28"}