Skip to content

Product observability: operability + diagnostics for core-runtime + core-cli (OSS scope) #797

Description

@DariuszNewecki

Recalibrated from the June maturity audit's internal observability items (Will fan-in/out metrics, shared-layer audit cadence, intent-read logging — monorepo-inward, dropped) to what matters for the two shipped products.

Scope

Unit Observability floor (OSS)
core-runtime Structured operational logging + health/readiness signals an operator needs to run it.
core-cli Actionable error / connection / auth diagnostics + a --verbose/debug mode — what a consumer sees when a command fails against a runtime.
boundary The core-cli ↔ core-runtime (OEM API) round-trip is observable/debuggable end-to-end.

Out of scope (deliberately)

The commercial Prometheus/SLA/incident-response bundle stays band-E Track 8 (sourcing:commercial, ADR-124). This issue is the OSS-appropriate floor: can someone operate and debug the two things we ship?

Part of the Production-Readiness last-mile epic. Directly serves external adoption.

Metadata

Metadata

Labels

priority:mediumDefault priority. Picked at session-start based on leverage and context, not urgency.type:taskSingle executable unit of work. One session, one outcome. The default for most issues.

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions