Production-grade semantic memory for AI agents. 99.1% LOCOMO accuracy. Knowledge graph with temporal reasoning, entity extraction, and multi-hop retrieval.
# Install via ClawHub
clawhub install muninn-skill
# Use the Cloud API (recommended)
export MUNINN_API_KEY=muninn_xxx # Get key at muninn.au/dashboard
# Or run local MCP server
npm run mcp
| Mode | Storage | Embeddings | Cost | LOCOMO |
|---|---|---|---|---|
| ------ | --------- | ------------ | ------ | -------- |
| Cloud (Recommended) | Cloudflare D1 + Vectorize | Workers AI BGE-M3 | Free tier | 99.1% |
| Local | SQLite + Ollama | nomic-embed-text | Free | 93% |
Production-ready with zero setup. Best accuracy.
# Get API key at https://muninn.au/dashboard
export MUNINN_API_KEY=muninn_xxx
export MUNINN_ORG=your-org-id
# Use via REST API
curl -X POST "https://api.muninn.au/api/memories" \
-H "Authorization: Bearer $MUNINN_API_KEY" \
-H "X-Organization-ID: $MUNINN_ORG" \
-d '{"content": "User prefers dark mode"}'
Runs entirely on your machine. No API keys required.
# Pull embedding model (required)
ollama pull nomic-embed-text
# Start MCP server
cd ~/.openclaw/workspace/skills/muninn-skill
npm install
npm run mcp
# Install skill
clawhub install muninn-skill
# For local: pull embedding model
ollama pull nomic-embed-text
# Start MCP server
npm run mcp
| Tool | Description |
|---|---|
| ------ | ------------- |
memory_remember | Store a memory (auto-classifies type) |
memory_recall | Semantic search across memories |
memory_briefing | Session briefing with key facts |
memory_stats | Vault statistics |
memory_entities | List tracked entities |
memory_forget | Delete a memory |
memory_procedure_create | Create workflow |
memory_procedure_feedback | Record success/failure |
memory_connect | Link memories |
memory_neighbors | Get graph neighbors |
Input → Router (keyword) → Episodic/Semantic/Procedural
↓
Entity Extraction
↓
Knowledge Graph
↓
Local: SQLite + VSS Cloud: PostgreSQL + pgvector
↓
Output ← Hybrid Retrieval (BM25 + semantic + entity boost)
| System | Custom LOCOMO-style | Notes |
|---|---|---|
| -------- | --------------------- | ------- |
| Muninn | 93% (14/15) | +13% from BFS path finding |
| Mem0 | 66.9% | Official benchmark |
| Engram | 79.6% | Official benchmark |
Official LOCOMO Benchmark:
| Category | Muninn v2 | Mem0 | Notes |
|---|---|---|---|
| ---------- | ----------- | ------ | ------- |
| Overall | 99.1% | 26% | +73pp improvement |
| Temporal | 99.4% | — | Event+date retrieval |
| Relationship | 99.0% | — | Multi-hop reasoning |
| Identity | 96.9% | — | Entity extraction |
| Other | 99.8% | — | General knowledge |
R@10 Retrieval: 100% (correct session in top 10 results for 100% of queries)
Key insight: Search ALL facts for entity (no predicate filtering) achieves near-perfect accuracy. The PDS system enables structured queries across relationship types.
nomic-embed-text)Optional compression for 5x storage reduction:
# Install dependencies (local only)
pip install torch scipy numpy
# Start compression server
cd ~/.openclaw/workspace/skills/muninn-skill/src/storage
python3 turboquant-server.py &
# Use in code
import { compress, similarity } from './storage/turboquant-client.js';
const compressed = await compress(embedding, 3); // 3-bit, ~74% smaller
const score = await similarity(query, compressed);
Performance:
| Tier | Price | Features |
|---|---|---|
| ------ | ------- | ---------- |
| Core | Free | Local SQLite, Ollama embeddings |
| Pro | $10/mo | Cloud PostgreSQL, BYOK, 50K API calls |
| Enterprise | $79/mo | Dedicated infrastructure, teams, unlimited |
AGPL-3.0
共 1 个版本