brainmd
Neuroplastic self-modifying runtime for AI agents. Creates a file-based 'brain' that learns from interactions: reflexes (fast-path responses), habits (learned patterns), weighted pathways (reinforcement), and a cortex (self-review loop). Use when: setting up adaptive agent behavior, creating learning loops, building persistent behavioral memory, or making an agent that improves over time.
npx skills add majiayu000/claude-skill-registry --skill brainmd --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# BRAIN.md *Self-modifying file system mimicking neuroplasticity for AI agents.* ## Concept brainmd gives an AI agent a file-based nervous system that strengthens successful behaviors, weakens failed ones, grows new pathways, and prunes unused ones. No separation between code and data — everything is mutable except the audit log. ## Quick Start 1. Initialize the brain structure: ```bash ./scripts/init-brain.sh [path] ``` 2. Seed initial pathways from observed behavior: ```bash node cortex/review.js record "reflex:my-pattern" true "Description of what worked" ``` 3. Run self-review (wire into heartbeat/cron): ```bash node cortex/review.js review ``` 4. Check neural status: ```bash node cortex/review.js status ``` ## Heartbeat Integration (Recommended) The key to making brainmd work is **automatic self-checks**. Without periodic review, the agent has to remember to use it — defeating the purpose. Add this to your agent's heartbeat/periodic routine (e.g. HEARTBEAT.md for OpenClaw): ```markdown ## 🧠 brainmd Self-Check (every heartbeat) Run cortex review and record any notable outcomes from recent interactions: node ~/.openclaw/workspace/brain/cortex/review.js review node ~/.openclaw/w
- Concept
- Quick Start
- Heartbeat Integration (Recommended)
- Architecture
- Core Mechanisms
- 1. Pathways (weights/pathways.json)
- 2. Reinforcement
- 3. Neurogenesis
- 4. Mutation Log
- 5. Reflexes
- 6. Habits
- 7. Cortex Integration
- Design Principles
- Bootstrapping Tips
node cortex/review.js record "reflex:my-pattern" true "Description of what worked" node cortex/review.js review node cortex/review.js status node cortex/review.js record "habit:some-behavior" true "What happened" node cortex/review.js record "habit:some-behavior" false "What went wrong" node cortex/review.js record "reflex:new-behavior" true "First time doing this" node reflexes/timing.js check 0.8 # Check with urgency=0.8
What does the brainmd skill do?
Neuroplastic self-modifying runtime for AI agents. Creates a file-based 'brain' that learns from interactions: reflexes (fast-path responses), habits (learned patterns), weighted pathways (reinforcement), and a cortex (self-review loop). Use when: setting up adaptive agent behavior, creating learning loops, building persistent behavioral memory, or making an agent that improves over time.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill brainmd --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.