AI Disclaimer

DiffGuardian uses AI models to assist with code review. Please read this before you rely on its output.

AI assistance is not a guarantee

DiffGuardian's reviews, summaries, and suggestions are generated by AI models and may be incomplete, inaccurate, or wrong. AI can miss real defects and can flag issues that are not real ("hallucinations"). Output is assistive only and is not a substitute for your own review, testing, and professional judgment.

You remain responsible

You are responsible for any code you ship and any decision you make with the help of DiffGuardian. Always review AI output before acting on it. Do not treat a passing DiffGuardian review as a warranty that code is correct, secure, or fit for any purpose.

Your AI provider governs your data

When you use your own AI provider keys (for example Anthropic or OpenAI), your code and prompts are sent directly from your machine to that provider under your agreement with them. Their terms and data-usage policies — not ours — apply to that processing. When you use a local model, your code stays on your machine. See our Security & Data Handling page for details.

No professional advice

DiffGuardian does not provide legal, security-compliance, or other professional advice. For high-stakes or regulated work, obtain appropriate independent review.

Changes

We may update this disclaimer as the product evolves. Material changes will be reflected by a new effective date and version.