Diffsift reviews the open pull requests in your Azure DevOps project or GitHub
repositories and turns each one into a structured review pack: findings with
severity, confidence, and evidence quoted from the actual diff, plus a risk
level, a suggested review focus, and questions for the author.
It runs on your own hardware with LM Studio, Ollama, or any OpenAI-compatible
endpoint, so code doesn't leave your machine. Hosted providers (OpenAI, Azure
OpenAI, Anthropic, OpenRouter) work too if your policies allow it, with a
confirmation before any code is sent.
Turn on monitoring and Diffsift reviews new and updated pull requests in the
background, so a first pass is waiting in its inbox when you sit down. It is
read-only: it never comments, approves, or merges. You copy the comments you
agree with.
No account, no telemetry. One-time license with a 14-day free trial.
No comments or reviews, maybe you want to be first?