Turbo Quant Memory MCP
A local-first Model Context Protocol server designed for AI coding agents. It provides compact retrieval and knowledge graph capabilities with project and global scopes. The server operates offline after initial model download, ensuring no data transmission, and supports hybrid search using BM25 and dense vectors.
Package and access
Local package with source-attributed installation metadata. Koskamo does not execute packages during review.
What you can do with it
- Manage local memory and knowledge graphs for coding agents
- Perform compact retrieval of project and global context
- Index code blocks and notes for semantic search
- Verify token savings via server_info()
Source and maintenance
- Maintenance checked
Packages and installation
Source-provided package data. Koskamo has not executed or live-probed this package.
turbo-quant-memory 0.34.0
{
"version": "0.34.0",
"transport": "stdio",
"identifier": "turbo-quant-memory",
"runtimeHint": "uvx",
"registryType": "pypi",
"registryBaseUrl": "https://pypi.org",
"packageArguments": [
{
"type": "positional",
"value": "serve",
"isRequired": true,
"description": "Run the memory server over stdio."
}
],
"environmentVariables": [
{
"name": "TQMEMORY_FTS_LANGUAGE",
"default": "English",
"description": "Snowball stemmer for the BM25 lane. Use Russian for Cyrillic-dominant corpora."
},
{
"name": "TQMEMORY_EMBEDDING_BACKEND",
"default": "fastembed",
"description": "Embedder backend. Default fastembed (ONNX); sentence-transformers needs the [torch] extra."
},
{
"name": "TQMEMORY_MIGRATE_ON_STARTUP",
"default": "0",
"description": "Set to 1 to apply pending schema migrations automatically on startup."
},
{
"name": "TQMEMORY_SECRETS_PASSPHRASE",
"description": "Argon2id passphrase for the secrets vault on headless hosts. Omit to use the OS keyring."
}
]
}Tools
Reported by the attributed source; not independently observed by Koskamo.
Connect from your client
Each snippet configures the same source-provided package; replace placeholders before use.
Cursor
{
"mcpServers": {
"turbo-quant-memory-mcp": {
"command": "uvx",
"args": [
"turbo-quant-memory==0.34.0",
"serve"
],
"env": {
"TQMEMORY_FTS_LANGUAGE": "<YOUR_TQMEMORY_FTS_LANGUAGE>",
"TQMEMORY_EMBEDDING_BACKEND": "<YOUR_TQMEMORY_EMBEDDING_BACKEND>",
"TQMEMORY_MIGRATE_ON_STARTUP": "<YOUR_TQMEMORY_MIGRATE_ON_STARTUP>",
"TQMEMORY_SECRETS_PASSPHRASE": "<YOUR_TQMEMORY_SECRETS_PASSPHRASE>"
}
}
}
}Cline
{
"mcpServers": {
"turbo-quant-memory-mcp": {
"command": "uvx",
"args": [
"turbo-quant-memory==0.34.0",
"serve"
],
"env": {
"TQMEMORY_FTS_LANGUAGE": "<YOUR_TQMEMORY_FTS_LANGUAGE>",
"TQMEMORY_EMBEDDING_BACKEND": "<YOUR_TQMEMORY_EMBEDDING_BACKEND>",
"TQMEMORY_MIGRATE_ON_STARTUP": "<YOUR_TQMEMORY_MIGRATE_ON_STARTUP>",
"TQMEMORY_SECRETS_PASSPHRASE": "<YOUR_TQMEMORY_SECRETS_PASSPHRASE>"
}
}
}
}VS Code
{
"servers": {
"turbo-quant-memory-mcp": {
"type": "stdio",
"command": "uvx",
"args": [
"turbo-quant-memory==0.34.0",
"serve"
],
"env": {
"TQMEMORY_FTS_LANGUAGE": "<YOUR_TQMEMORY_FTS_LANGUAGE>",
"TQMEMORY_EMBEDDING_BACKEND": "<YOUR_TQMEMORY_EMBEDDING_BACKEND>",
"TQMEMORY_MIGRATE_ON_STARTUP": "<YOUR_TQMEMORY_MIGRATE_ON_STARTUP>",
"TQMEMORY_SECRETS_PASSPHRASE": "<YOUR_TQMEMORY_SECRETS_PASSPHRASE>"
}
}
}
}Claude Desktop
{
"mcpServers": {
"turbo-quant-memory-mcp": {
"command": "uvx",
"args": [
"turbo-quant-memory==0.34.0",
"serve"
],
"env": {
"TQMEMORY_FTS_LANGUAGE": "<YOUR_TQMEMORY_FTS_LANGUAGE>",
"TQMEMORY_EMBEDDING_BACKEND": "<YOUR_TQMEMORY_EMBEDDING_BACKEND>",
"TQMEMORY_MIGRATE_ON_STARTUP": "<YOUR_TQMEMORY_MIGRATE_ON_STARTUP>",
"TQMEMORY_SECRETS_PASSPHRASE": "<YOUR_TQMEMORY_SECRETS_PASSPHRASE>"
}
}
}
}Let your agent find its tools
Find servers like Turbo Quant Memory MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.