regent-vcs

regent-vcs / re_gent

Public

Git for AI coding agents.

re-gent.dev ai-agent claude-code developer-tools devtools golang
98
2
100% credibility
Found May 08, 2026 at 98 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Go
AI Summary

re_gent is a tracking system that records AI agent actions on code, providing history logs, line-by-line attribution, and conversation context like a specialized version control for AI workflows.

How It Works

1
📰 Discover re_gent

You learn about a handy tool that tracks every change your AI coding helper makes, like a diary for your project.

2
📥 Install easily

You grab the tool with a simple download, ready to use in moments.

3
🏗️ Set up your project

You prepare your workspace with one quick action, and tracking begins automatically.

4
🤖 Work with your AI

Your AI assistant edits code, runs commands, and chats as usual – everything gets captured without extra effort.

5
📖 Review the story

You peek at the full history to see steps, tools used, and conversations behind each change.

6
🔍 Trace a code line

You check any line to discover exactly which AI instruction wrote it and when.

🎉 Master your AI changes

You now have a complete record to rewind errors, understand decisions, and confidently build with AI.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 98 to 98 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 re_gent?

re_gent brings Git-style version control to AI coding agents, tracking every tool call like edits, writes, or bash commands in your codebase. Built in Go, it auto-captures workspace snapshots via hooks in tools like Claude Code, storing them in .regent/ for git github repo audits. Run rgt log for step history, rgt blame file:line to trace which prompt or session wrote specific code.

Why is it gaining traction?

It solves the "AI broke my code, now what?" pain with per-line blame to prompts, concurrent session branches, and conversation tracking that outlives chat clears—features regular Git lacks for ai coding github flows. No manual commits needed; just init and work. The VS Code extension delivers inline annotations, making git coding style provenance instant.

Who should use this?

Claude Code users or teams with github coding agents refactoring codebases, needing git github difference clarity on AI changes. Ideal for devs facing git coding questions like "which prompt added this bug?" or enforcing git coding guidelines in collaborative AI-assisted git github gitlab workflows.

Verdict

Worth installing via Homebrew for AI-heavy projects—core commands work reliably in active development. But with 98 stars and 1.0% credibility, it's POC-stage; hold for v1 rewind and GC if you need battle-tested stability.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.