stella-ops.org/git.stella-ops.org
1
0
Fork 0
Code Issues Pull Requests Actions 2 Releases Wiki Activity

Refine unified search answer shaping and viability

Browse Source
This commit is contained in:
master
2026-03-07 21:49:10 +02:00
parent 8f43378317
commit bbfa27ca39
15 changed files with 719 additions and 35 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
docs/implplan/SPRINT_20260307_033_AdvisoryAI_search_query_understanding_and_viability.md
View File

@@ -21,7 +21,7 @@
## Delivery Tracker
### AI-ZL2-001 - Remove mode-shaped answer assumptions from unified search
Status: TODO
Status: DONE
Dependency: none
Owners: Developer
Task description:
@@ -34,7 +34,7 @@ Completion criteria:
- [ ] Clarify and insufficient fallbacks remain deterministic.
### AI-ZL2-002 - Strengthen suggestion viability and corpus readiness signals
Status: TODO
Status: DONE
Dependency: AI-ZL2-001
Owners: Developer
Task description:
@@ -47,7 +47,7 @@ Completion criteria:
- [ ] Integration tests cover live-readiness and empty-corpus behavior.
### AI-ZL2-003 - Keep telemetry optional and non-blocking
Status: TODO
Status: DONE
Dependency: AI-ZL2-002
Owners: Developer
Task description:
@@ -63,6 +63,8 @@ Completion criteria:
| Date (UTC) | Update | Owner |
| --- | --- | --- |
| 2026-03-07 | Sprint created for the backend half of the operator correction pass on unified search. | Project Manager |
| 2026-03-07 | Implementation started: tightening query-driven answer shaping, corpus-readiness-aware suggestion viability, and telemetry-disabled parity before rerunning backend and Playwright suggestion suites. | Developer |
| 2026-03-07 | Completed backend query-driven answer shaping, grounded-only suggestion viability, and telemetry-off parity. Evidence: `dotnet build src/AdvisoryAI/__Tests/StellaOps.AdvisoryAI.Tests/StellaOps.AdvisoryAI.Tests.csproj -v minimal -p:BuildInParallel=false -p:UseSharedCompilation=false`; `StellaOps.AdvisoryAI.Tests.exe -class StellaOps.AdvisoryAI.Tests.UnifiedSearch.UnifiedSearchServiceTests`; `StellaOps.AdvisoryAI.Tests.exe -class StellaOps.AdvisoryAI.Tests.Integration.UnifiedSearchEndpointsIntegrationTests`; `StellaOps.AdvisoryAI.Tests.exe -class StellaOps.AdvisoryAI.Tests.UnifiedSearch.SearchAnalyticsServiceTests`; `StellaOps.AdvisoryAI.Tests.exe -class StellaOps.AdvisoryAI.Tests.UnifiedSearch.SearchQualityMonitorTests`; live query `database connectivity` returned `contextAnswer.status = grounded`. | Developer |
## Decisions & Risks
- Decision: FE should not be responsible for choosing answer mode or rescuing dead suggestions.