Over the past few months, I've been focused on reducing risks from AI by improving the evaluations that tell us what dangers exist in AI models. To support this effort, I've contributed to Inspect AI over the past month+ (November 8, 2025 - December 10, 2025) in the following ways:

• Added a CONTRIBUTING guide to welcome and orient new people
• Updated the README to show how to build docs
• Improved documentation by fixing errors in the CTF tutorial
• Fixed broken links in the docs
• Added .qmd to .editorconfig to specify rules for QMD files
• Opened an issue to get "good first issues" labeled
• Worked to close several issues to make it easier for people to find issues that actually need to be done

I plan to continue providing support via critical documentation, project management, and code to make it easier for people to contribute to Inspect. This ultimately leads to more and better evaluations for the orgs that use it -- Anthropic, Google DeepMind, Apollo Research, SecureBio, and more.

I'd love to have your support as I work to make AI safer through my contributions to Inspect.