Mcp Victoriametrics
Let your coding agent query VictoriaMetrics metrics, run PromQL-style exploration, and pull official docs while you debug production or design dashboards.
Overview
VictoriaMetrics MCP (mcp-victoriametrics) is an Operate-phase MCP server that connects coding agents to your VictoriaMetrics API and documentation for live metrics investigation.
What is this MCP server?
- Stdio MCP server (v1.20.2) with OCI image ghcr.io/victoriametrics/mcp-victoriametrics
- Connects to vmsingle or vmselect via VM_INSTANCE_ENTRYPOINT plus bearer token auth
- Supports single vs cluster instance types for correct API routing
- Optional VictoriaMetrics Cloud integration via VMC_API_KEY
- Documentation-aware tooling alongside VictoriaMetrics HTTP API
- Server version 1.20.2
- OCI package ghcr.io/victoriametrics/mcp-victoriametrics:1.20.2
- Required env: VM_INSTANCE_ENTRYPOINT, VM_INSTANCE_TYPE, VM_INSTANCE_BEARER_TOKEN
Community signal: 178 GitHub stars.
What problem does it solve?
Jumping between agent chat, PromQL docs, and your metrics endpoint slows down solo operators who need fast answers during incidents or tuning.
Who is it for?
Indie builders running VictoriaMetrics (self-hosted or Cloud) who want agent-assisted metric lookup and API exploration during operate and monitoring work.
Skip if: Teams on Datadog-only stacks with no VictoriaMetrics endpoint, or anyone who only needs static dashboard screenshots without API access.
What do I get? / Deliverables
After you register the server with your agent and env vars, the assistant can query VictoriaMetrics and reference docs in one thread while you fix or improve monitoring.
- Agent-callable VictoriaMetrics API and documentation tools
- Repeatable stdio MCP wiring with version 1.20.2 OCI image
- Faster metric investigation loops during incidents
Recommended MCP Servers
Journey fit
Production observability and incident response sit in Operate; this MCP is the live bridge to your metrics store. Monitoring is the canonical shelf for time-series query and VictoriaMetrics API access—not generic infra provisioning.
How it compares
MCP integration to VictoriaMetrics APIs, not an embedded metrics database or a Grafana plugin.
Common Questions / FAQ
Who is mcp-victoriametrics for?
It is for developers and solo operators who already use VictoriaMetrics and want their AI agent to call the metrics API and docs safely from the IDE.
When should I use mcp-victoriametrics?
Use it during Operate and monitoring when you are debugging alerts, validating queries, or learning VictoriaMetrics behavior alongside production data.
How do I add mcp-victoriametrics to my agent?
Add the MCP server entry (stdio or your client’s OCI pull for ghcr.io/victoriametrics/mcp-victoriametrics:1.20.2), set VM_INSTANCE_ENTRYPOINT, VM_INSTANCE_TYPE, and VM_INSTANCE_BEARER_TOKEN, then restart the agent.