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

experiment: validate behaviour selection against real assessments

Browse Source
This commit is contained in:
robbond
2026-08-06 16:30:10 +01:00
parent 51648e4b8f
commit da3d35f437
4 changed files with 563 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/current-project-state.md
+3 -5
View File
@@ -46,7 +46,7 @@ These are what currently affect the working engine:
The following were built during Experiments 18–25B. They are isolated diagnostic layers with no active integration into the user-facing investigation:
- Investigation-state assessment (phase and progress classification);
- Behaviour selection from assessed state;
- Behaviour selection from assessed state — passively evaluated in Experiment 39 against real assessment outputs; produces all valid behaviours but Acknowledge dominates when health=healthy;
- Decision condition status evaluation;
- Question-to-condition relevance scoring;
- Evidence direction classification (support, contradict, inform);
@@ -75,7 +75,7 @@ The following were built during Experiments 18–25B. They are isolated diagnost
## 6. Work Currently Paused
- Engine experiments are paused after Experiment 25B;
- Engine experiments are at Experiment 39 (passive Behaviour Selection validation, awaiting review);
- UI experiments are paused;
- Knowledge-management experiments are complete (confirmed by Experiment 38 cold-start validation);
- Nothing historical has been deleted or archived yet.
@@ -100,12 +100,10 @@ Historical documents are retained under `docs/archive/` and should be opened onl
## 8. Return-to-Work Summary
Engine experiments are paused after Experiment 25B, which established scope-aware condition status — distinguishing direct evidence from relevant-but-different claims by checking subject, timeframe and claim type. The passive classifier layers (Exps 18–25B) remain isolated; none yet control the user-facing investigation. Current focus is knowledge management: creating a short current-state entry point so future sessions do not need to read the full experiment history.
Engine experiments advanced to Experiment 39, a passive validation of Behaviour Selection against real Investigation State Assessment outputs. The existing selector produces all valid behaviours but Acknowledge dominates (71%) because it fires first when health=healthy. No user-facing engine behaviour changed. Knowledge-management phase is complete: all seven criteria met. No backlog items deleted or promoted.
First document to read: `docs/current-project-state.md`. Then consult `.claude/architecture-guardrails.md` before any code changes and `docs/project-knowledge-inventory.md` for task-specific references. The full experiment history remains available in `docs/design-evolution-log.md` but is no longer default reading.
Engine experiments remain paused after Experiment 25B. UI experiments are paused. Knowledge-management phase is complete: all seven criteria met. No backlog items deleted or promoted. **First file to inspect when resuming:** `docs/current-project-state.md`.
## Verification Marker
Implementation status last checked against source: Experiment 28.