Skip to main content

/tsh-implement-observability

info

Not invoked directly by users. To trigger observability implementation, use /tsh-implement — the Engineering Manager will automatically delegate to the DevOps Engineer.

Agent: DevOps Engineer File: .github/internal-prompts/tsh-implement-observability.prompt.md

Implements comprehensive observability solutions covering metrics, logs, traces, and alerting.

How It’s Triggered​

/tsh-implement <describe what to monitor or observe>

The Engineering Manager identifies observability tasks in the plan and delegates them to the DevOps Engineer automatically.

What It Does​

1. Context Discovery​

  • Identifies existing monitoring stack (Prometheus, Grafana, Datadog, CloudWatch, etc.).
  • Checks for existing dashboards, alert rules, and log aggregation patterns.
  • Discovers service level objectives (SLOs) and key performance indicators (KPIs).

2. Implementation​

  • Sets up metrics collection with appropriate instrumentation.
  • Configures log aggregation with structured logging patterns.
  • Implements distributed tracing across services.
  • Creates alerting rules with appropriate thresholds and escalation paths.
  • Builds dashboards for system health visibility.

3. Verification​

  • Validates that metrics are being collected correctly.
  • Ensures alert rules fire as expected with test scenarios.
  • Verifies trace propagation across service boundaries.

Skills Loaded​

  • tsh-implementing-observability — Observability patterns for logging, monitoring, alerting, tracing.
  • tsh-technical-context-discovering — Project conventions and existing patterns.