- Add --rag-debug flag to show retrieved chunk names and similarity scores
- Add explicit fallback notices when RAG indexing/query embedding fails
- Log RAG index/query metrics (duration, scores, top hit, token estimate)
- Normalize and cap chunk content for more stable prompt shape on small models
- Add hypothesis-continuity instruction for follow-up prompts
- Add retrieval scoring API and new tests for truncation/fallback/debug paths
- Replace plain 'tai>' prompt with styled console.input() bold cyan prompt
- Wrap interactive mode entry in a Rich Panel with border
- Frame each AI response with Rule dividers (──── AI Response ────)
- Style guardrail warnings with ⚠ prefix and bold yellow
- Improve /help output with formatted Panel showing all commands
- Style collection report: ✓/✗ per item with color, truncation in dim
- Style probe output: ✓/✗ with green/red, host info in dim
- Add Rule header divider on session start
