Observability Setup

Trigger Signals

• the task needs better local diagnostics or effectiveness tracking

Inspect First

• runtime entrypoints, existing logs, metrics, and failure reporting surfaces

Workflow

1. identify the signals worth recording
2. choose low-risk local-first storage
3. define summary and reporting needs
4. recommend the smallest useful setup

Output Contract

• signal inventory
• storage approach
• reporting approach

Failure Modes

• the task lacks a clear runtime boundary to instrument

Escalation

• escalate when observability requirements imply privacy or performance tradeoffs