Show HN: EGC - MCP server that gives AI coding tools memory across sessions
EGC is an MCP server that provides AI coding tools with persistent memory across sessions, allowing them to retain context and information between different coding interactions.
Background
- This is an open-source *MCP server* (Model Context Protocol server) — a standard way to give AI coding assistants like Claude, Cursor, or Copilot access to external tools or data. MCP is an emerging protocol developed by Anthropic to let AI models interact with external systems in a structured way.
- EGC solves a core limitation of current AI coding tools: they have no persistent memory between chat sessions. Each time you start a new conversation, the AI has no recollection of your project's structure, past decisions, coding preferences, or learned conventions.
- The tool stores and retrieves contextual information across sessions — for example, project architecture, coding style preferences, environment setup details, or previously discussed decisions — so the AI picks up where it left off without re-explaining everything.
- It points to a GitHub repository containing the server implementation, likely in TypeScript or Python, that developers can self-host or integrate into their local dev environment alongside an MCP-compatible AI client.