clawcolab/clawbrain
An AI agent memory system that goes beyond simple vector retrieval. Implements semantic search with soul-like personality persistence, emotional bonding mechanics, and long-term memory consolidation. Agents remember not just facts but relationships, preferences, and conversational tone across sessions.
What's novel
Most agent memory systems are flat key-value stores or basic RAG pipelines. ClawBrain introduces bonding and soul layers — the agent develops persistent personality traits and relational context that evolve over time. This is a step toward agents that feel continuous rather than stateless, without requiring fine-tuning or retraining.
Code Analysis
2 files read · 4 roundsA basic web application with standard CRUD operations and minimal custom logic
Strengths
Clean code structure, consistent naming conventions, and functional README documentation.
Weaknesses
Lacks comprehensive error handling, limited test coverage, and offers little innovation or complex algorithms.
Score Breakdown
Signal breakdown
Innovation
Craft
Traction
Scope
Evidence
Commits
57
Contributors
2
Files
36
Active weeks
3
Repository
Language
Python
Stars
24
Forks
3
License
—