Description
Rank 1 - high value, low implementation cost. The two live write-stack usage datasources expose 325 metric names, but a positive rate, an active gauge and a billing-period quantity mean different things. Build bounded per-stack views for product producing signals from the existing write-stack reader, with explicit windows and absent-not-zero behavior. Proposed new emitted series: 0; point-in-time views and direct panels suffice.
Acceptance Criteria
- #1 Each product signal has a documented metric-specific window, unit and nonzero interpretation
- #2 Views distinguish missing input from measured zero and never infer human UI use
- #3 No raw metric names or stack identifiers become new metric labels
Definition of Done
- #1 just test
- #2 just tf-validate
- #3 just check-identifiers and just no-em-dashes both return clean
Implementation Notes
Parked at Wave 1 boundary: researched build candidate in doc-0006, outside this wave implementation scope. Resume when Rob selects the next product work; first verify the candidate metric or datasource contract named in this task.
Wave 2 code landed at 6ecd68b: coverage_producing_signals is a point-in-time view of documented Metrics active series and Traces bytes per second, each using the capability-adoption 24-hour peak. Missing, measured zero and positive are distinct; no human UI use is inferred. No metrics or labels were added. CodeRabbit completed with zero findings; targeted checks 217 passed, 2 skipped. Final exact-SHA gate, CI and dev readback pending.
Final Summary
Code acceptance: two documented backend signals, missing versus measured-zero semantics, no new metric series; reviewed and tested at product 51ce670 with Gate G and hosted CI 35935370824 success. Dev live view publication was not proved: T2 failed before scan on the consumer projection mismatch and dev was rolled back. Do not cite code acceptance as live proof.