LGTM PR Review installs statistics

LGTM is an AI code reviewer that automatically analyzes every pull request and posts findings as review comments, bugs, security issues, and performance problems surfaced before your PR is merged.

Powered by Claude. Bring your own API key.

How It Works

Install LGTM, add your ANTHROPIC_API_KEY to your repository secrets, and every pull request gets automatically reviewed.

When a PR is opened or updated, LGTM:

• Analyzes the diff using Claude AI • Identifies bugs, security vulnerabilities, and performance issues • Posts a structured review comment on the PR with all findings

What It Detects

• Bugs — logic errors, off-by-one issues, null pointer risks • Security — injection vulnerabilities, missing auth checks, unsafe operations • Performance — inefficient patterns, unnecessary allocations • Code quality — anti-patterns and maintainability issues

Each finding includes:

• Severity level (critical / high / medium / low) • Confidence score • Evidence from the code • A concrete suggestion for fixing it

Setup

• Install LGTM from GitHub Marketplace • Add ANTHROPIC_API_KEY to your repository or organization secrets • Open a pull request — LGTM reviews it automatically

Pricing

Free. You pay only for your own Anthropic API usage (typically $0.01–0.10 per review, depending on diff size).

Open Source

LGTM is fully open source. View the code, report issues, or contribute at github.com/KeatonShawhan/LGTM.