Frequently Asked Questions (FAQ) about MCP Server Memo

Q: What is the MCP Server Memo? A: The MCP Server Memo is a lightweight server that implements the Model Context Protocol (MCP) for managing session memory for Large Language Models (LLMs). It allows you to store, retrieve, and manage conversation history for your AI agents.

Q: What is MCP (Model Context Protocol)? A: MCP is an open protocol that standardizes how applications provide context to LLMs, enabling them to access external data sources and tools.

Q: How does the MCP Server Memo work? A: The server uses the local filesystem to store session summaries and their history. It provides MCP tools for creating, retrieving, listing, updating, and appending summaries.

Q: What are the key features of MCP Server Memo? A: Key features include persistent history (all historical versions are preserved), time-ordered organization of sessions, local storage (no external database required), MCP compliance, performance optimization, and minimal dependencies.

Q: What MCP tools are provided by the server? A: The server provides the following MCP tools: upsertSummary, getSummaryTool, listSummariesTool, updateMetadata, appendSummary, listAllSummariesTool, and getSessionHistory.

Q: What is sessionId and how should I use it? A: sessionId is a unique identifier for a conversation session. The client application is responsible for generating this ID once at the beginning of a new logical conversation session, preferably using a UUID (Version 4) library. The same ID must be reused for all subsequent calls related to that session.

Q: How do I install MCP Server Memo? A: You can install it by cloning the repository, installing dependencies using npm install, and building the project with npm run build.

Q: How do I configure the server? A: The server can be configured using environment variables, such as MCP_SUMMARY_DIR, to specify the directory for storing summaries.

Q: How do I run the server? A: You can run the server using npm start for standard startup, npm run dev for development mode with auto-reload, or npm run start:log to start with logging to a file.

Q: Can I use an external database instead of the local filesystem? A: Currently, the MCP Server Memo is designed to use the local filesystem. Support for external databases is not yet available but may be considered for future releases.

Q: How does appendSummary work? A: The appendSummary tool reads the existing latest summary (if any), adds two newlines, appends the new content, and saves it as a new version. All previous versions are preserved.

Q: What is UBOS and how does it relate to MCP Server Memo? A: UBOS is a full-stack AI Agent Development Platform that helps you orchestrate AI Agents, connect them with your enterprise data, build custom AI Agents with your LLM model, and create Multi-Agent Systems. The MCP Server Memo can be used as an asset within the UBOS platform to enhance the context management capabilities of your AI Agents.

Q: Where can I find more information about the MCP protocol? A: You can find more information about the Model Context Protocol (MCP) on the official MCP website or related documentation.