DevFlow connects your GitHub account to a structured AI review pipeline. Open a pull request, generate a review, and get back a plain-language summary, a ranked list of likely bugs with file locations and severity scores, and actionable suggestions - all in a sticky panel alongside the diff.
Reviews are stored per-user. Come back to them later. Regenerate when the code changes. Everything is scoped to you with row-level security.
|
|
|
|
|
|
| Framework | Next.js 16 · App Router · Turbopack |
| Language | TypeScript strict |
| Styling | Tailwind CSS v4 |
| Auth + DB | Supabase - GitHub OAuth, Postgres, RLS |
| GitHub | Octokit REST |
| AI | Gemini gemini-3.5-flash · JSON-schema-constrained output |
| Animation | Framer Motion |
| Typography | Fraunces (display serif) + Geist Sans/Mono |
You'll need: Node 20+, a Supabase project with GitHub OAuth enabled, and a Gemini API key.
1. Install dependencies.
npm install2. Set up your environment.
cp .env.example .env.local| Variable | Source |
|---|---|
SUPABASE_URL |
Supabase -> Project Settings -> API |
SUPABASE_ANON_KEY |
Supabase -> Project Settings -> API |
SUPABASE_SERVICE_ROLE_KEY |
Supabase -> Project Settings -> API |
GEMINI_API_KEY |
Google AI Studio |
GITHUB_PROVIDER_TOKEN_COOKIE_SECRET |
Any random 32-char string |
NEXT_PUBLIC_SUPABASE_URL |
Same as SUPABASE_URL |
NEXT_PUBLIC_SUPABASE_ANON_KEY |
Same as SUPABASE_ANON_KEY |
3. Apply the database schema - run supabase/migrations/20260524000000_init.sql in the Supabase SQL Editor.
4. Start the dev server.
npm run devOpen localhost:3000 and sign in with GitHub.
Built with Next.js · Supabase · Gemini · MIT





