Alan-Jowett / CoPilot-For-Consensus / 20340116186

78%

push

github

Add observability metrics to orchestrator's summary deduplication logic (#389)

* Initial plan

* Implement retrieval set comparison in orchestrator to prevent unnecessary summaries

Co-authored-by: Alan-Jowett <20480683+Alan-Jowett@users.noreply.github.com>

* Add warning when thread document not found during retrieval set update

Co-authored-by: Alan-Jowett <20480683+Alan-Jowett@users.noreply.github.com>

* Merge main branch and add observability metrics to summary_id-based deduplication

Resolved merge conflict by adopting main branch's cleaner summary_id-based approach (calculates hash from thread+chunks to check summaries collection) instead of storing retrieval_set in thread documents.

Enhanced main's implementation by adding:
- orchestrator_summary_skipped_total metric (reason: summary_already_exists)
- orchestrator_summary_triggered_total metric (reason: chunks_changed)
- Updated tests to match new summary_id approach

This preserves the optimization benefits while avoiding schema changes and leveraging existing data structures.

Co-authored-by: Alan-Jowett <20480683+Alan-Jowett@users.noreply.github.com>

* Document summary deduplication logic and metrics in orchestrator README

Updated README.md to reflect:
- Orchestration flow pseudo-code now shows summary_id calculation and deduplication check
- Added documentation for new metrics (orchestrator_summary_skipped_total, orchestrator_summary_triggered_total) with their reasons

Addresses code review feedback about missing documentation.

Co-authored-by: Alan-Jowett <20480683+Alan-Jowett@users.noreply.github.com>

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: Alan-Jowett <20480683+Alan-Jowett@users.noreply.github.com>

4 of 4 new or added lines in 1 file covered. (100.0%)

4459 of 5381 relevant lines covered (82.87%)

0.93 hits per line