memory_search

MCP tool from Membot by project-you-apps

Multi-signal search: cosine + Hamming + keyword reranking

How to use it

Project documentation lists memory_search for the Membot MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), then check which tools your installed version makes available. Tool availability can depend on configuration and credentials. A server handshake does not verify this tool’s behavior. See the full listing for setup details.

Install Membot

$python /path/to/membot/membot_server.py
FULL MEMBOT LISTING

Other tools in Membot (17)

federate_consolidate

Find cross-machine matches across mounted carts, write a consolidated cart

federate_load

Mount every machine's federated cart in a fleet directory at once

federate_migrate_jsonl

One-time migration from JSONL learning logs to brain carts

federate_publish

Append session learning entries to a machine's federated cart

get_passage

Retrieve a specific passage by index (for navigation and drill-down)

get_status

Server diagnostics (mounted cartridge, memory count, GPU status)

list_cartridges

Browse available brain cartridges with size and capabilities

memory_store

Store new text into the mounted cartridge

mount_cartridge

Load a cartridge into memory (embeddings + optional GPU brain)

multi_list

List every cart currently mounted in the pool

multi_mount

Add a cart to the multi-cart pool with optional cartid and role

multi_mount_directory

Mount every cart in a directory matching a glob pattern (used by federation)

multi_search

Search across mounted carts with scope and rolefilter, results attributed to source

multi_unmount

Remove a cart from the pool by cartid

passage_links

Get navigation links (prev/next/parent/related) for a passage

save_cartridge

Persist the current cartridge to disk (secure NPZ format)

unmount

Free memory and unload the current cartridge