Memwright provides AI agents with persistent, searchable memory that remains outside the context window until needed, allowing for efficient retrieval and management of memories.
From the registry: Embedded memory for AI agents with SQLite, pgvector, and Neo4j graph search.
$ uv tool install memwrightPlease install the `memwright` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install memwright** — Install from PyPI Run: `uv tool install memwright`
Canonical MCP server config (stdio transport):
- command: `memwright`
- args: ["mcp"]
- optional environment variables:
- `MEMWRIGHT_PATH`: Default store path (example: `~/.memwright`)
- `MEMWRIGHT_URL`: Remote API URL (distributed mode)
- `MEMWRIGHT_NAMESPACE`: Default namespace (example: `default`)
- `MEMWRIGHT_TOKEN_BUDGET`: Default token budget (example: `16000`)
- `MEMWRIGHT_SESSION_ID`: Session ID for provenance tracking
Note: Also installable via pipx, pip, or poetry. Downloads ~90MB embedding model on first use.
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.AI Agents Framework with Self Reflection and MCP support