Regent
Version control for AI coding agents — audit, undo, and replay every session.
Visit website →About
Regent brings version control to AI coding agents. Where Git tracks file changes, Regent tracks agent actions, every prompt, edit, and decision, stored in a .regent/ directory alongside the code.
Each conversation gets its own branch. A developer can undo specific agent changes, trace which prompt caused a particular line, replay past sessions, and run multiple agents in parallel without conflicts.
It works with Claude Code, Codex, and OpenCode out of the box, is free and open source under Apache 2.0, and is in public alpha. It is aimed at developers who want an audit trail and undo for agent-driven changes, not just the resulting file diffs.
AI-assisted draft, human-reviewed before publishing — see how we choose & review tools.
Why we picked it
Regent made the cut for a genuinely new idea — Git tracks file changes, but tracing which specific prompt caused a specific line, and being able to replay a past session, is worth naming as version control for the decision, not just the diff.
Related tools
- 1 Tiden Verifies coding agents built what you asked for — specs as code Trial
- 2 GitGlow Local Git workbench with smart review for your changes and agent edits Freemium
- 3 GitHub Copilot AI coding assistant inside your editor, with a free tier Freemium
- 4 Together Link Run frontier open models in the coding harness you already use Paid
- 5 Octri Turns OpenAPI specs into docs, SDKs in 10 languages, and MCP server Freemium
- 6 FastRouter.ai One OpenAI-compatible API to route across 200+ LLMs with smart failover Freemium