Gentleman-Programming

Gentleman-Programming / engram

Public

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

603
47
100% credibility
Found Feb 17, 2026 at 155 stars 4x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Go
AI Summary

Engram is a standalone program that lets AI coding assistants store and retrieve structured memories from past sessions using a local file for persistence, with easy viewing and team sharing options.

How It Works

1
📖 Discover Engram

You hear about Engram, a simple helper that gives your AI coding assistant a lasting memory so it doesn't forget past work.

2
🛠️ Get it set up

Grab the ready-to-use program from a trusted spot like your computer's package manager or download page – it takes just seconds.

3
🔗 Link to your AI tool

Tell your AI coding buddy to use Engram by adding a quick note in its settings, like pointing it to a helpful friend.

4
✨ AI starts remembering

As you code together, your AI automatically saves key moments like bug fixes, smart choices, and lessons learned into its personal notebook.

5
🖥️ Browse your memories

Open a beautiful, easy-to-navigate screen in your terminal to search, read details, and relive past sessions at a glance.

6
📂 Share with your team

Drop memory files into your shared project folder so everyone’s AI can pull in the same history when working together.

🚀 Work flows perfectly

Next time you start, your AI instantly recalls everything from before, picking up right where you left off and saving hours.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 155 to 603 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is engram?

Engram is a persistent memory system for AI coding agents, letting tools like Claude Code, OpenCode, Cursor, or Windsurf remember decisions, bug fixes, and session summaries across restarts via structured MCP tools such as mem_save and mem_search. Built as a single Go binary with SQLite and FTS5 full-text search, it exposes a CLI (engram tui, search, sync), HTTP API, and terminal UI—no Node.js, Docker, or vector DBs needed. Developers get agent-curated, searchable engrams that survive compactions, with git-friendly sync for teams.

Why is it gaining traction?

Unlike claude-mem's Claude-only focus and multi-runtime deps, engram is agent-agnostic, runs one binary via stdio MCP, and trusts the agent (already LLM-powered) for clean summaries—no noisy tool logs or extra API calls. Git sync exports compressed chunks to .engram/ for conflict-free repo sharing, auto-importing on clone. The TUI dashboard, timeline views, and compaction hooks make memory feel seamless, standing out amid github persistent storage projects like iptables-persistent or persistent homology github.

Who should use this?

AI agent power users on long projects—Claude Code devs debugging auth flows, OpenCode teams tracking architecture shifts, or Cursor users recalling N+1 fixes. Ideal for solo coders or small teams needing shared engrams without setup overhead, especially if you're tired of re-explaining context each session.

Verdict

Grab it via Homebrew for quick wins on MCP agents; 109 stars and solid docs show promise despite 1.0% credibility signaling early maturity. Test on a side project—its zero-deps purity nails persistent memory where others bloat.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.