Monitoring and observability MCP servers for logs, metrics, traces and error tracking: let an AI assistant investigate incidents with real data. 699 servers in this category.
Proxies Gemini Vertex AI completions with OpenTelemetry trace spans to provide observability for LLM responses.
$claude mcp add --transport http traced-llm-proxy https://anthropic-mcp-opentelemetry-api-264025.getvda.ai/mcp
Verifies that AI agent execution traces contain the necessary telemetry for post-hoc causal reconstruction of failure classes.
$claude mcp add spec -- uvx evidence-readiness
Provides security, observability, and governance for AI agents by auditing tool calls and enforcing allow/block policies locally.
Enables AI clients to monitor and manage Veeam Backup & Replication, Veeam ONE, and Veeam Service Provider Console.
Provides read-only tools to monitor Redis instances, including queue depths, Celery status, connected clients, and server metadata.
Enables AI clients to manage Zabbix Enterprise Monitoring through 100 tools for network discovery, device onboarding, and professional reporting.
Streams real-time context to AI agents via Kafka and WebSockets for live anomaly detection and autonomous root-cause diagnosis.
Allows AI agents to bypass GitHub's 14-day limit by backing up, analyzing, and querying historical repository traffic metrics via a local SQLite database.
Provides read-only access to Graylog for searching logs, analyzing alerts, monitoring cluster health, and investigating incidents via MCP tools.
Allows AI agents to access Grafana dashboards, datasources, Prometheus/Loki queries, alert rules, incidents, and Sift investigations via a secure gateway.
Lets an AI client monitor and manage Unraid servers via the official GraphQL API with read-only defaults and opt-in mutations.
Lets AI coding assistants analyze Windows WPR/ETW traces by loading .etl files and answering natural language questions about performance data.
Audit the token consumption of your MCP servers to understand how much context window they occupy before your agent uses them.
Enables AI clients to investigate incidents by reading data from Kubernetes, Grafana, Prometheus, Slack, and Argo CD.
Enables small local LLMs to query Zabbix 8.0 monitoring data using plain language for metrics, host status, and problem history.
Lets an AI assistant monitor real-time security incidents, explain risk scores, and query transaction activity through a voice-queryable interface.
Provides AI agents with incident context by integrating PagerDuty, Dynatrace, and operational runbooks for automated triage and remediation.
Tracks undocumented AWS changes and links them to EKS incidents to automate cloud infrastructure auditing and triage.
Provides local observability, usage analytics, and error debugging for AI coding agents and local LLM services via a SQLite-backed dashboard.
Records MCP host interactions, including model inputs, tool choices, and MCP operations, into a portable SQLite database for debugging and verification.