Shodh-Memory is a persistent memory server for AI agents and robots, exposed through MCP, HTTP, Docker, and Zenoh/ROS2. It stores, recalls, and organizes memories and tasks locally with usage-based decay and association so context persists across sessions.
From the registry: Cognitive memory for AI agents — semantic search, Hebbian learning, knowledge graphs.

Please install the `shodh-memory` MCP server into my current AI client (that's you).
Canonical MCP server config (stdio transport):
- command: `npx`
- args: ["-y","@shodh/memory-mcp"]
- optional environment variables:
- `SHODH_API_KEY`: API key (auto-generated for local use; required for remote server) (example: `<your-api-key>`)
Note: 37 tools: persistent memory with Hebbian learning, decay, spreading activation. No LLM calls needed, <200ms store. Also covers GTD todos, projects, reminders. Rust binary auto-downloaded. Supports robotics/ROS2 via Zenoh.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.SHODH_API_KEYrequiredAPI key required for remote server access to /api endpoints and MCP client connectionsMCP server for searching Airweave collections with natural language queries.