ainative-zerodb-memory-mcp

Persistent Memory for AI Agents

Optimized MCP server providing 6 focused tools for agent memory management with advanced context window optimization, semantic search, and cross-session memory.

Before: Monolithic server with 77 tools consuming 10,400+ tokens After: Focused server with 6 tools consuming ~800 tokens Result: 92% reduction in context footprint, faster agent decisions, better accuracy

• Automatic token limiting - Never exceed LLM context windows
• Intelligent pruning - Keep important and recent memories
• Memory decay - Old memories naturally fade over time
• Importance scoring - Automatically rank memory significance

• Vector embeddings - BAAI BGE models (384, 768, 1024 dimensions)
• Semantic search - Find by meaning, not just keywords
• Cross-session memory - Remember across conversations
• Auto-embedding - No manual embedding required

• ZeroLocal - localhost:8000 (fast, free, private)
• ZeroDB Cloud - api.ainative.studio (scalable, managed)
• Auto-detection - Automatically finds available endpoint

# Clone repository
git clone https://github.com/ainative/zerodb-memory-mcp.git
cd zerodb-memory-mcp

# Install dependencies
npm install

# Configure environment
cp .env.example .env
# Edit .env with your credentials

# Test locally
npm start

export ZERODB_API_URL="http://localhost:8000"  # or cloud URL
export ZERODB_USERNAME="your-email@example.com"
export ZERODB_PASSWORD="your-password"
export ZERODB_PROJECT_ID="your-project-id"

{
  "mcpServers": {
    "zerodb-memory": {
      "command": "node",
      "args": ["/path/to/zerodb-memory-mcp/index.js"],
      "env": {
        "ZERODB_API_URL": "http://localhos