github

In an effort to combat AI slop code reviews in open source projects Mitchell Hashimoto has created Vouch.

People must be vouched for before interacting with certain parts of a project (the exact parts are configurable to the project to enforce). People can also be explicitly denounced to block them from interacting with the project.

The tool is being tested in Ghostty and comes with CLI and GitHub action integrations. The tool generates and uses a "single flat file", so it's easily parsable.

This would allow projects to vouch or denounce users and share their lists with others.

Vouch lists can also form a web of trust. You can configure Vouch to read other project's lists of vouched or denounced users. This way, projects with shared values can share their trust decisions with each other and create a larger, more comprehensive web of trust across the ecosystem.

Terence Eden recently commented on GitHub's community discussion on the same AI slop issue. The suggestions there were around building a users reputation score around account age, labelling based on other GitHub user interactions and restricting contributions to assigned issues, among others. These were obviously centred around GitHub specific implementations unlike Vouch which would work across hosted git providers.

I discovered Vouch through Simon Willison's weblog.

CommonMark is the Markdown specification created by John MacFarlane, Jeff Atwood and others, to encompass the various flavours of Markdown that was adopted by different software over the years. GitHub adopted CommonMark along with it's extension for Markdown called GitHub Flavored Markdown (GFM) sometimes around 2016-2017. I've complained about how different platforms deveate from the standard. This GitHub Engineering post shows how good of a job the CommonMark contributors did to represent common usage of Markdown with only 1% deveating from GitHub's previous Markdown parser.

We [GitHub] actually enabled CommonMark for all new user comments in the website several months ago, with barely anybody noticing — this is a testament to the CommonMark team’s fantastic job at formally specifying the Markdown language in a way that is representative of its real world usage.

All in all, less than 1% of the input documents were modified by the normalization process, matching our [GitHub's] expectations and again proving that the CommonMark spec really represents the real-world usage of the language.