autogame-17

autogame-17 / evolver

Public

The GEP-Powered Self-Evolution Engine for AI Agents. Genome Evolution Protocol. | evomap.ai

evomap.ai
1,097
129
89% credibility
Found Feb 04, 2026 at 17 stars 65x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

A self-improvement engine for AI agents that analyzes runtime history to generate protocol-guided prompts for safe, auditable enhancements.

How It Works

1
🔍 Discover the Helper

You hear about a smart helper that makes your AI assistant get better on its own by learning from its past mistakes.

2
📥 Bring It Home

You grab the simple package and place it in your AI's workspace, ready to start helping right away.

3
🚀 Start the Magic Scan

With one easy click, you tell it to go, and it quietly reviews your AI's recent chats and notes to spot ways to improve.

4
💭 Get Smart Suggestions

It creates a clear guide with ideas on fixes or upgrades, like a personalized plan for your AI to grow stronger.

5
Choose Your Path
👀
Review First

Check the suggestions yourself and approve before any changes happen.

✅
Trust and Go

Let it make the safe updates automatically with built-in checks.

6
🧪 Test the Changes

Run a quick check to see your AI working smoother, with fewer hiccups and better results.

🎉 Smarter AI Achieved

Your assistant evolves into a more reliable helper, learning and improving over time with full records of its growth.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 1,097 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 evolver?

Evolver is a JavaScript engine for self-evolving AI agents on GitHub, implementing the Genome Evolution Protocol (GEP) to analyze runtime logs, extract error signals or opportunities, and output structured prompts for safe improvements. It turns chaotic prompt tweaks into auditable Genes and Capsules—reusable assets for repair, optimization, or innovation—without auto-editing core code. Fire up `node index.js` for instant GEP-guided evolution, or `--loop` for continuous self-repair in agent setups like OpenClaw.

Why is it gaining traction?

Unlike free-form self-evolving agents GitHub projects, evolver enforces protocol constraints with strategies like `repair-only` or `innovate`, plus signal de-duplication to avoid loops. A2A sharing lets nodes exchange Capsules across agents, echoing self-evolving agents papers and surveys. Developers dig the ops toolkit—lifecycle management, skill monitoring, protected files—for reliable, zero-dependency evolution.

Who should use this?

AI agent maintainers scaling prompt governance in production loops, OpenClaw skill builders hardening flaky behaviors, or teams auditing evolutions for compliance. Ideal for backend devs wrestling log-driven fixes, not one-off hackers skipping history.

Verdict

Grab it for agent evolver experiments (331 stars shows real interest), with strong docs, CLI ops, and MIT license offsetting the 0.8999999761581421% credibility score—early maturity means sandbox first, but ops resilience shines.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.