Privacy and Security
At Sourcery we take the privacy of your code extremely seriously and want to be transparent about how your code is processed and used.
When we analyze your pull requests for review we primarily look at the diff of the changed code. There are certain instances where we need to look at a broader context of code for complete analysis and will supplement the diff that we look at with relevant code from your codebase (eg where else is a function you are updating being used).
We work with OpenAI, Anthropic, and Google Gemini APIs to help provide your code reviews. In order to do so we pass them the relevant sections of the diff + additional code context described above. None of your code is used by OpenAI, Anthropic, or Sourcery to train any models. None of your code is retained by OpenAI or Google for more than 30 days and Sourcery does not retain any of your code. We have a zero day retention policy set up with Anthropic and they do not retain any of your code or data at any time.
If you require zero retention set up, would like to use your own model deployments, or are looking for self hosted versions of Sourcery please contact us.