home / mcp / memory bank mcp server
Memory Bank is an MCP server that helps teams create, manage, and access structured project documentation. It generates and maintains a set of interconnected Markdown documents that capture different aspects of project knowledge, from high-level goals to technical details and day-to-day progress.
Configuration
View docs{
"mcpServers": {
"tuncer-byte-memory-bank-mcp": {
"url": "https://mcp.memorybank.example.com/mcp",
"headers": {
"GEMINI_API_KEY": "your_gemini_api_key_here"
}
}
}
}Memory Bank MCP is an MCP server that provides structured project knowledge to LLM-powered agents and tools. It generates and maintains interconnected Markdown documents across a project, enabling teams to access, update, and search knowledge via MCP-compatible clients.
Connect Memory Bank MCP to an MCP client (such as an LLM agent or IDE integration) using the MCP workflow. You launch Memory Bank MCP as a local server, then point your MCP client to it or run it as a stdio component within your MCP orchestration. You can initialize a Memory Bank structure, update documents, run queries across all documents, and export the knowledge base for sharing or backup. Your client can request document content, trigger AI-assisted updates, or perform context-aware searches to retrieve relevant project information.
Prerequisites: make sure you have Node.js installed on your machine. You can verify with node -v and npm -v.
1. Clone the Memory Bank MCP repository.
2. Install dependencies.
3. Set up your Gemini API key if you plan to generate documentation via Gemini.
Memory Bank MCP is intended to run as an MCP server. You typically launch it as part of an MCP workflow, and connect to it from an MCP-compatible client.
Keep Gemini API keys secret. Store keys in a secure environment and avoid committing them to version control. Use least-privilege access and rotate keys as needed.
If you encounter connection issues, verify that Memory Bank MCP is running and that your MCP client is configured to connect to the correct command or URL. Check that environment variables (such as the Gemini API key) are correctly set in the running process.
Creates a new Memory Bank structure with all document templates and an initial project goal.
Updates a specific document in the Memory Bank and optionally regenerates it using AI.
Searches across all documents with context-aware relevance ranking to find relevant information.
Exports all Memory Bank documents in a chosen format for backup or sharing.