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

experiment(confidence-engine): add openai provider apparatus

Browse Source
This commit is contained in:
robbond
2026-09-06 07:50:47 +01:00
parent 653934559c
commit 860ee6fc5b
3 changed files with 133 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/current-handoff.md
+6
View File
@@ -116,6 +116,12 @@
- This increment adds only a final semantic-preservation self-check, testing instruction salience rather than missing semantic specification. No live model calls were made.
- Next boundary: one manual fixed-scenario Postman production run, scoring C first while checking A/B/D1/D2/E and unsupported expansion or causal strengthening for regressions.
## OpenAI reconstruction-provider experiment apparatus
- Structural Ollama/provider issues were previously fixed; current semantic comparison now warrants testing model-compliance variance.
- An experiment-only OpenAI Responses API provider apparatus exists for `gpt-5.6-terra`, using the same current v0.5 prompt and canonical reconstruction schema. Production provider selection remains Ollama/Qwen.
- No live OpenAI calls have occurred. Next boundary: a bounded live semantic comparison, not production migration.
## Current product architecture
Three distinct routes, not a single page: