Roborev

Roborev

โ˜… 3.8 ๐Ÿ”– 1200 saves ๐Ÿ’ฐ Free Code Review TestingGitHub

Roborev is an open-source, local-first code review daemon built for the age of AI coding agents. It has 1,769 stars on GitHub and is under active development in 2026. Roborev reviews every commit in the background, using the agent tools you already run, so problems surface in seconds instead of at pull-request time.

Overview

Coding agents can land dozens of commits an hour, far faster than humans read code. Roborev installs a git post-commit hook that queues a background review of each commit through agents such as Codex, Claude Code, Gemini, or Copilot, all on your own machine. Findings go into a persistent ledger with severities and file locations, and an agent hook feeds open findings back into the active session so the agent fixes its own work while context is fresh. Bundled fix and refine skills can iterate a branch until every review passes. Nothing is hosted: reviews run through your existing agent subscriptions and keys. Every review stores the exact prompt and the agent log alongside its verdict, so a finding can be audited later instead of trusted blindly. Optional PostgreSQL sync lets a team’s machines converge on one shared review history while each daemon keeps working fully locally.

Key Features

  • Automatic background review of every commit via a git post-commit hook
  • Works with 11+ agent CLIs, auto-detected, with per-workflow model routing and failover
  • Persistent findings ledger in a terminal UI and browser UI; nothing closes itself
  • Multi-reviewer panels that synthesize one verdict, plus security and design review types
  • Pull request polling for GitHub and GitLab with synthesized review comments
  • Single Go binary with SQLite storage; no per-seat pricing and no code leaves your network

๐Ÿ’ณ Pricing

Roborev is free and open source under the MIT license. It orchestrates the agent CLIs and subscriptions you already pay for, and adds no hosted service fees of its own.

Who Is It For

Roborev is for developers and teams leaning heavily on coding agents who want continuous, accountable review of agent-written code before it ever reaches a human reviewer.

๐Ÿš€ Visit Roborev Official Site

๐Ÿšฉ Report an issue