# MCP production monitoring checklist

Updated October 16, 2026. By Flowlines.

Use this checklist for one server and one representative user journey, then repeat after deployment.

1. Define one useful user job and an observable completion condition.
2. Establish health probes and auth/protocol failure monitoring for your transport and revision.
3. Instrument tool executions once and verify error classification and duration units.
4. Review identity, content capture, redaction and retention with synthetic sensitive values.
5. Check context propagation, export delivery and graceful shutdown; keep stdio stdout clean.
6. Segment by server, environment, release and tool; keep unknown clients and users visible.
7. Test success, tool error, timeout, recovery and repeated-call journeys.
8. Define alert windows, minimum volumes, owners, runbooks and notification destinations.
9. Configure a useful product or business signal and review its false positives.
10. Check telemetry and outcome coverage before comparing adoption or release impact.

## Acceptance checks

- A successful tool call, tool-level error and timeout appear with distinct classifications.
- One execution is not counted twice by overlapping instrumentation.
- Sample secrets do not reach any telemetry destination.
- Uncorrelated calls and unknown outcomes remain explicit.
- Low traffic and absent telemetry are not presented as healthy success rates.
- A synthetic alert reaches the intended owner; recovery clears it.
- Business signals are reviewed as findings, not treated as confirmed revenue.

## Start with Flowlines

Connect Flowlines or install the official plugin. Ask to onboard your MCP, approve the OpenTelemetry plan, then ask to check onboarding. No proprietary runtime SDK or additional tracing platform is required.

- [Full monitoring guide](https://flowlines.ai/blog/mcp-observability-guide)
- [Onboarding](https://flowlines.ai/docs)
- [Official plugin](https://github.com/flowlines-ai/plugins)
- [Example Prometheus rules](https://flowlines.ai/examples/mcp-monitoring/alerts.yml)
- [Synthetic rule tests](https://flowlines.ai/examples/mcp-monitoring/alerts.test.yml)

The rules use custom normalized metrics, not official OTel metric names. Adapt and validate them before use. Monitoring your server requires instrumentation; connecting an assistant alone does not install it.