Split out of the #80/#95 fix (commit a9d1d24, branch fix/scene-context-leak-80-95).
a9d1d24 closed the scene-context leak for embedding and severity in incident_correlator._build_context (same shape as the #87 coords leak): each scene is now judged on its own value, not the primary scene's copy on the call doc.
Still leaking:llm_correlator.py:63-66_call_block reads ctx["call_doc"]'s whole-calltranscript for every scene. That text is the input to both the cheap correlation tier and the tiebreaker, so on a multi-scene call every scene's LLM decision is made against a transcript that includes the other scenes. Same defect class as #87/#80/#95, third field.
Fix shape: thread the scene's own transcript (the segment slice already available in the scene loop) into ctx and have _call_block read that instead of call_doc["transcript"].
Also: with a9d1d24 more scenes now qualify thin, and llm_correlator.py:227 skips the LLM for thin calls, so the LLM tier's measured share drops again — #96's re-measurement of #35 must postdate this fix too.
Found by drb-correlation-review while reviewing a9d1d24.
Split out of the #80/#95 fix (commit `a9d1d24`, branch `fix/scene-context-leak-80-95`).
`a9d1d24` closed the scene-context leak for `embedding` and `severity` in `incident_correlator._build_context` (same shape as the #87 coords leak): each scene is now judged on its own value, not the primary scene's copy on the call doc.
**Still leaking:** `llm_correlator.py:63-66` `_call_block` reads `ctx["call_doc"]`'s **whole-call** `transcript` for every scene. That text is the input to both the cheap correlation tier and the tiebreaker, so on a multi-scene call every scene's LLM decision is made against a transcript that includes the other scenes. Same defect class as #87/#80/#95, third field.
**Fix shape:** thread the scene's own transcript (the segment slice already available in the scene loop) into `ctx` and have `_call_block` read that instead of `call_doc["transcript"]`.
**Also:** with `a9d1d24` more scenes now qualify thin, and `llm_correlator.py:227` skips the LLM for thin calls, so the LLM tier's measured share drops again — #96's re-measurement of #35 must postdate this fix too.
Found by `drb-correlation-review` while reviewing `a9d1d24`.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Split out of the #80/#95 fix (commit
a9d1d24, branchfix/scene-context-leak-80-95).a9d1d24closed the scene-context leak forembeddingandseverityinincident_correlator._build_context(same shape as the #87 coords leak): each scene is now judged on its own value, not the primary scene's copy on the call doc.Still leaking:
llm_correlator.py:63-66_call_blockreadsctx["call_doc"]'s whole-calltranscriptfor every scene. That text is the input to both the cheap correlation tier and the tiebreaker, so on a multi-scene call every scene's LLM decision is made against a transcript that includes the other scenes. Same defect class as #87/#80/#95, third field.Fix shape: thread the scene's own transcript (the segment slice already available in the scene loop) into
ctxand have_call_blockread that instead ofcall_doc["transcript"].Also: with
a9d1d24more scenes now qualify thin, andllm_correlator.py:227skips the LLM for thin calls, so the LLM tier's measured share drops again — #96's re-measurement of #35 must postdate this fix too.Found by
drb-correlation-reviewwhile reviewinga9d1d24.