- name
- engramai
- description
- Neuroscience-grounded memory for AI agents. Add, recall, and manage memories with ACT-R activation, Hebbian learning, and cognitive consolidation.
- homepage
- https://github.com/tonitangpotato/neuromemory-ai
- metadata
- {"clawdbot":{"emoji":"🧠","requires":{"bins":["python3"],"packages":{"pip":["engramai"]}}}}
engramai 🧠
Cognitive memory system implementing ACT-R activation, Memory Chain consolidation, Ebbinghaus forgetting, and Hebbian learning.
Installation
pip install engramaiQuick Start
from engram import Memory
mem = Memory("./agent.db")
mem.add("User prefers concise answers", type="relational", importance=0.8)
results = mem.recall("user preferences", limit=5)
mem.consolidate() # Daily maintenanceCLI Usage
# Add a memory
neuromem add "User prefers dark mode" --type preference --importance 0.8
# Recall memories
neuromem recall "user preferences"
# View statistics
neuromem stats
# Run consolidation (like sleep)
neuromem consolidate
# Prune weak memories
neuromem forget --threshold 0.01
# List memories
neuromem list --limit 20
# Show Hebbian links
neuromem hebbian "dark mode"AI Agent Integration (Important!)
For AI agents to use engram correctly, follow these patterns:
When to Call What
| Trigger | Action | Example |
|---|---|---|
| Learn user preference | store(type="relational") | "User prefers concise answers" |
| Learn important fact | store(type="factual") | "Project uses Python 3.12" |
| Learn how to do something | store(type="procedural") | "Deploy requires running tests first" |
| Question about history | recall() first, then answer | "What did I say about X?" |
| User satisfied | reward("positive feedback") | Strengthens recent memories |
| User unsatisfied | reward("negative feedback") | Suppresses recent memories |
| Daily maintenance | consolidate() + forget() | Run via cron or heartbeat |
What to Store
✅ Store:
- User preferences and habits
- Important facts and decisions
- Lessons learned
- Procedural knowledge
❌ Don't store:
- Every conversation message (too noisy)
- Temporary information
- Publicly available facts
- Sensitive data (unless requested)
Importance Guide
| Level | Use For |
|---|---|
| 0.9-1.0 | Critical info (API keys location, absolute preferences) |
| 0.7-0.8 | Important (code style, project structure) |
| 0.5-0.6 | Normal (general facts, experiences) |
| 0.3-0.4 | Low priority (casual chat, temp notes) |
Hybrid Mode (Recommended)
Use engram alongside file-based memory:
- engram: Active memory — retrieval, associations, dynamic weighting
- **Files (memory/*.md)**: Logs — transparency, debugging, manual editing
Heartbeat Maintenance
Add to your heartbeat or cron:
## Memory Maintenance (Daily)
- [ ] engram.consolidate
- [ ] engram.forget --threshold 0.01Memory Types
factual— Facts and knowledgeepisodic— Events and experiencesrelational— Relationships and preferencesemotional— Emotional momentsprocedural— How-to knowledgeopinion— Beliefs and opinions
MCP Server
For Claude/Cursor/Clawdbot integration:
python -m engram.mcp_server --db ./agent.dbMCP Config (Clawdbot):
mcp:
servers:
engram:
command: python3
args: ["-m", "engram.mcp_server"]
env:
ENGRAM_DB_PATH: ~/.clawdbot/agents/main/memory.dbTools: engram.store, engram.recall, engram.consolidate, engram.forget, engram.reward, engram.stats, engram.export
Key Features
| Feature | Description |
|---|---|
| ACT-R Activation | Retrieval ranked by recency × frequency × context |
| Memory Chain | Dual-system consolidation (working → core) |
| Ebbinghaus Forgetting | Natural decay with spaced repetition |
| Hebbian Learning | "Neurons that fire together wire together" |
| Confidence Scoring | Metacognitive monitoring |
| Reward Learning | User feedback shapes memory |
| Zero Dependencies | Pure Python stdlib + SQLite |
Links
- PyPI: https://pypi.org/project/engramai/
- npm: https://www.npmjs.com/package/neuromemory-ai
- GitHub: https://github.com/tonitangpotato/neuromemory-ai
- Docs: https://github.com/tonitangpotato/neuromemory-ai/blob/main/docs/USAGE.md