Agentscreator

Agentscreator / Engram

Public

Multi-agent memory consistency platform. We're hiring contributors—check HIRING.md

www.engram-us.com agent-memory collaboration conflict-resolution consistency distributed-systems
40
36
100% credibility
Found Apr 09, 2026 at 40 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Engram creates shared, persistent memory for teams of AI coding agents that detects contradictions and ensures knowledge consistency across sessions.

How It Works

1
🔍 Discover Engram

You hear about Engram, a helpful way for your team's AI coding helpers to share important discoveries without forgetting them between chats.

2
📥 Install with one command

Copy and paste a simple command into your terminal, and it quietly connects Engram to your favorite coding app like Cursor or VS Code.

3
🔄 Restart your coding app

Close and reopen your coding app, and everything is ready without any extra setup.

4
Tell your AI to connect
✨
Create new team space

Your AI makes a fresh shared memory space and gives you a simple invite code to share.

🔑
Join teammate's space

Paste the invite code from a teammate, and you're instantly connected.

5
🧠 AI helpers share memories

Now every AI on your team remembers key facts, spots disagreements, and builds on each other's work seamlessly.

6
📊 Check the dashboard

Visit a simple web page to see all shared knowledge, review conflicts, and track your team's AI activity.

🎉 Team AI works smarter together

Your coding team saves time as AIs avoid repeating mistakes and stay consistent across projects.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 40 to 40 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 Python-based shared memory platform for AI agent teams, letting agents commit verified facts that persist across sessions and instantly query team knowledge via MCP tools like engram_commit and engram_query. It detects contradictions between agents—using entity matching, NLI models, and numeric rules—and surfaces them for resolution, all while keeping data encrypted and workspace-isolated. Developers get a dashboard to visualize memory graphs, conflicts, and agent activity without managing databases.

Why is it gaining traction?

In a world of fragmented LLM multi-agent memory setups, Engram stands out with one-command installers that auto-configure 20+ MCP clients like Claude, Cursor, VS Code Copilot, and Zed, enabling zero-setup multi-agent coding GitHub workflows. Privacy is ironclad—no data read, shared, or trained on—plus research-backed features like forgetting-by-design and conflict tiers address real pain in multi-agent memory management from a computer architecture perspective. Early adopters praise the invite-key sharing for team onboarding.

Who should use this?

Engineering teams running parallel AI agents in IDEs—think backend devs coordinating Claude and Copilot on microservices, or full-stack squads debugging distributed systems where one agent uncovers rate limits the others miss. Ideal for GitHub Copilot multi-agent systems or DeepSeek-powered workflows needing consistent facts without manual syncing.

Verdict

Try it if you're building multi-agent GitHub flows; the MCP integration and conflict detection deliver immediate value despite 40 stars and 1.0% credibility signaling alpha maturity. Run the installer, join a workspace, and query—docs are solid, but expect rough edges in scaling.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.