jonathasrochadesouza

Revai Platform — AI skill for Claude Code

AI community

Local-first AI code review desktop app - run AI code review on your machine with your own agent harness (Claude Code, Codex, Kiro CLI, OpenCode, Copilot).

How to install Revai Platform

This entry records only its repository, not the path inside it, so there is no exact command to give. Open jonathasrochadesouza/revai-platform and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Revai Platform does

Local-first AI code review desktop app - run AI code review on your machine with your own agent harness (Claude Code, Codex, Kiro CLI, OpenCode, Copilot). Windows, macOS & Linux.

Alternatives in AI

  • Open Science — Open Science Desktop — local-first, model-agnostic AI research workbench for macOS, Windows & Linux 1.5k ★
  • The Pair — Open-source AI pair programming for desktop: a Mentor + Executor agent cross-check each other's code to catch 358 ★
  • Agentpet — A desktop pet for macOS & Windows that monitors your AI coding agents (Claude Code, Codex, Cursor, Gemini 334 ★

README

Revai Platform — local-first AI code review

Revai Platform

**AI code review, running 100% on your machine using your harness.**

[![status](https://img.shields.io/badge/status-mvp-facc15?labelColor=18181b)](#-getting-started) [![release](https://img.shields.io/github/v/release/jonathasrochadesouza/revai-platform?color=38bdf8&labelColor=18181b)](https://github.com/jonathasrochadesouza/revai-platform/releases/latest) [![license](https://img.shields.io/badge/license-MIT-4ade80?labelColor=18181b)](LICENSE) [![node](https://img.shields.io/badge/node-%E2%89%A520.9-334155?labelColor=18181b)](https://nodejs.org) [![privacy](https://img.shields.io/badge/privacy-local_data-4ade80?labelColor=18181b)](#-data--privacy)


**Revai Platform** is a free, open-source **desktop app for local-first AI code review** on Windows, macOS and Linux. You point it at a GitHub/GitLab repository (branch, PR, MR, commit) **or a local folder**, and the AI coding agent already on your machine — Claude Code, Codex, Kiro CLI, OpenCode or GitHub Copilot — runs the review through a full UI: 5-step pipeline, Quality Gate, findings triage and an automated fix loop. Your code, diffs and rules never leave your computer; the platform is the local host for the [`revai` skill](https://github.com/jonathasrochadesouza/revai).

✨ Highlights

  • 👀 5 steps live — resolve → clone → diff → skill → report, with a streaming log (SSE)
  • ✅ Quality Gate — faithful mirror of computeGate(): fails only with an open critical
  • 🔁 Fix loop — applies the fix, runs the tests and shows the Gate delta
  • 🤖 Harness of your choice — kiro-cli, claude, codex, opencode or copilot, configured in onboarding
  • 💾 100% local — no telemetry; data in ~/.revai, commits only with your click

🧭 How it works

The pipeline is the skill's own (`revai.mjs` + `generate-review.mjs`); the platform orchestrates and renders it: