MemFlow MCP
Enables Large Language Models to store and retrieve persistent memories with intelligent search capabilities.
memflow-mcp
A Model Context Protocol (MCP) server that enables Large Language Models to store and retrieve persistent memories with intelligent search capabilities.
Description
MemFlow MCP provides seamless integration between LLMs like Claude and your Memory Bank API, allowing for persistent memory management across conversations. The server supports adding memories with tags, semantic search, and flexible memory retrieval.
Installation
npx memflow-mcp
Or install globally:
pnpm add -g memflow-mcp
Configuration
Environment Variables
MEMBANK_API_URL=http://localhost:3000
MEMBANK_API_KEY=your-api-key # optional
Claude Desktop
See CLAUDE_DESKTOP_CONFIG.md for detailed configuration options for different Node.js installations.
Quick start - Find your npx path and use it:
which npx
Then configure Claude Desktop:
{
"mcpServers": {
"memflow": {
"command": "/your/npx/path",
"args": ["-y", "memflow-mcp"],
"env": {
"MEMBANK_API_URL": "http://localhost:3000"
}
}
}
}
Available Tools
- addMemory - Store content with optional tags
- searchMemory - Search memories with semantic matching
- listMemories - Browse stored memories with filtering
Usage
Once configured, you can use these commands in Claude:
Add this to memory: "Claude can now remember things across conversations"
Search my memories for "conversations"
List my recent memories
Requirements
- Node.js 18+
- Memory Bank API server running
Troubleshooting
"spawn npx ENOENT" Error
-
Find your npx path:
which npxUse this full path in your Claude Desktop config.
-
Common paths:
- asdf:
/Users/username/.asdf/shims/npx - Homebrew (Intel):
/usr/local/bin/npx - Homebrew (Apple Silicon):
/opt/homebrew/bin/npx - System:
/usr/bin/npx
- asdf:
-
Test it works:
/your/npx/path -y memflow-mcp
"Unexpected token" JSON Errors
- Ensure
MEMBANK_API_URLis set correctly - Check that your Memory Bank API is running
- Restart Claude Desktop after config changes
API Endpoints
Your Memory Bank API should support:
POST /memory- Create memoryGET /memory- List/search memories
License
MIT
相關伺服器
Apache Gravitino
Access Apache Gravitino, a high-performance, federated metadata lake for data and AI.
ODBC MCP Server
Enables LLM tools to query databases using ODBC connections.
ClickHouse
An MCP server for interacting with a ClickHouse database.
Unity Catalog MCP Server
An MCP server that allows LLM agents to seamlessly execute functions within Unity Catalog.
Dremio
Integrate Large Language Models (LLMs) with Dremio, a data lakehouse platform.
Open Formula 1 MCP Server
MCP Server to retrieve and analyze Formula 1 (F1) races, cars, drivers, lap, pit stops and more.
Pinterest by CData
A read-only MCP server for querying live Pinterest data, powered by the CData JDBC Driver.
CData AlloyDB MCP Server
A read-only MCP server for AlloyDB, enabling LLMs to query live data directly from AlloyDB databases.
MCP Snowflake Reader
Read-only access to Snowflake databases. Requires Snowflake connection information provided via MCP client configuration.
Kollect MCP Server (Standalone)
An MCP server for querying data from the Kollect tool's API endpoints, such as /api/data, /api/snapshots, and /api/costs.