Engram

Active
GitHub Go MIT

Description

Engram is a persistent memory system for AI coding agents. Agent-agnostic Go binary with SQLite + FTS5, MCP server, HTTP API, CLI, and TUI interfaces.

Key Features

  • Persistent memory system for AI coding agents with SQLite + FTS5 full-text search
  • Agent-agnostic: works with Claude Code, OpenCode, Gemini CLI, Codex, VS Code, Cursor, and any MCP-compatible agent
  • Single Go binary with zero dependencies — no Node.js, Python, or Docker required
  • 19 MCP tools covering save, search, session lifecycle, conflict detection, and utilities
  • Interactive TUI dashboard with vim-style navigation and Catppuccin theme
  • Git-based sync for sharing memories across machines with compressed chunks

Use Cases

💡 Giving coding agents persistent memory across sessions so they remember past decisions and context
💡 Sharing team knowledge and architecture decisions through cloud-synced memory
💡 Debugging complex multi-session coding projects with searchable memory timeline
💡 Managing memory hygiene with session summaries, compaction, and conflict surfacing

Tags

memory sqlite mcp coding-agent

Categories

🧩 Agent Memory

Quick Start

1. Install: brew install gentleman-programming/tap/engram. 2. Setup your agent: engram setup claude-code (or opencode, gemini-cli, codex). 3. Agent calls mem_save to store memories; next session searches via mem_search for relevant context.
Visit GitHub

Related Projects