Skip to main content
Glama

MCP Memory Service

claude_desktop_config_template.json969 B
{ "mcpServers": { "memory": { "_comment": "Recommended: Use Python module approach (most stable, no path dependencies)", "command": "python", "args": [ "-m", "mcp_memory_service.server" ], "_alternative_approaches": [ "Option 1 (UV): command='uv', args=['--directory', '${PROJECT_PATH}', 'run', 'memory', 'server']", "Option 2 (New script path): command='python', args=['${PROJECT_PATH}/scripts/server/run_memory_server.py']", "Option 3 (Legacy, shows migration notice): command='python', args=['${PROJECT_PATH}/scripts/run_memory_server.py']" ], "env": { "MCP_MEMORY_STORAGE_BACKEND": "sqlite_vec", "MCP_MEMORY_CHROMA_PATH": "${USER_DATA_PATH}/mcp-memory/chroma_db", "MCP_MEMORY_BACKUPS_PATH": "${USER_DATA_PATH}/mcp-memory/backups", "PYTORCH_ENABLE_MPS_FALLBACK": "1", "PYTORCH_CUDA_ALLOC_CONF": "max_split_size_mb:128" } } } }

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/doobidoo/mcp-memory-service'

If you have feedback or need assistance with the MCP directory API, please join our Discord server