Felipe Cardoso
|
997cfaa03a
|
feat(memory): implement memory reflection service (#99)
Add reflection layer for memory system with pattern detection, success/failure
factor analysis, anomaly detection, and insights generation. Enables agents to
learn from past experiences and identify optimization opportunities.
Key components:
- Pattern detection: recurring success/failure, action sequences, temporal, efficiency
- Factor analysis: action, context, timing, resource, preceding state factors
- Anomaly detection: unusual duration, token usage, failure rates, action patterns
- Insight generation: optimization, warning, learning, recommendation, trend insights
Also fixes pre-existing timezone issues in test_types.py (datetime.now() -> datetime.now(UTC)).
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
|
2026-01-05 04:22:23 +01:00 |
|