robbond/confidence-engine
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(confidence-engine): honor openai alternate output schema

Browse Source
This commit is contained in:
robbond
2026-09-07 10:18:41 +01:00
parent bb3082d633
commit 0eadef6e3b
3 changed files with 66 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/current-handoff.md
+6
View File
@@ -33,6 +33,12 @@ If YES, the next live experiment is one timed/costed OpenAI UI investigation mea
- Browser request contracts and client state remain unchanged. Deterministic coverage includes initial start, normal update, episode reconsideration, focused deconstruction, overview synthesis, and Current Understanding synthesis.
- Zero live calls occurred. Next boundary: one real Playwright-driven Terra investigation measuring user-visible latency, actual LLM-call sequence, and OpenAI usage/cost.
## OpenAI alternate structured output
- The OpenAI provider now honors a caller-supplied structured-output schema, applying its existing strict-schema transport projection; absent an alternate schema, initial reconstruction retains its existing strict schema and transport normalization.
- This deterministically fixes the focused-deconstruction 502 root cause: its six-field schema now reaches OpenAI rather than being replaced by the initial-reconstruction schema. Zero live calls occurred.
- Next boundary: retry one fresh real Playwright-driven OpenAI/Terra investigation; do not reuse the failed focused-deconstruction attempt.
## Repository checkpoint
- **Branch:** `feature/initial-decomposition-v0.61`