prism-review

AI-powered GitHub PR code review CLI. Fetches PR data (diff, file contents, comments) and sends it to OpenAI for a thorough code review, with an interactive follow-up session.

Setup

Prerequisites

• Python 3.10+
• uv (recommended) or pip
• A GitHub personal access token
• An OpenAI API key

Installation

git clone https://github.com/your-username/prism-review.git
cd prism-review
uv venv && source .venv/bin/activate
uv pip install -e .

Configuration

Set the required environment variables:

export GITHUB_TOKEN=ghp_your_token_here
export OPENAI_API_KEY=sk-your_key_here

Or copy .env.example to .env and fill in the values.

Usage

prism review <pr-url-or-owner/repo> [pr_number] [--model MODEL] [--no-interactive]

Examples

# Review a PR by URL
prism review https://github.com/facebook/react/pull/42

# Or by owner/repo + number
prism review facebook/react 42

# Use a specific model
prism review https://github.com/django/django/pull/18231 --model gpt-4.1

# Skip interactive follow-up
prism review https://github.com/torvalds/linux/pull/999 --no-interactive

Options

Option	Default	Description
--model	gpt-5.4	OpenAI model to use
--no-interactive	false	Skip interactive follow-up session

How It Works

1. Fetches PR metadata, diff, changed file contents, and comments via the GitHub API
2. Sends the assembled context to OpenAI with a senior code reviewer system prompt
3. Renders the review as Markdown in the terminal
4. Enters an interactive session for follow-up questions (type quit or exit to stop)

License

MIT