Just released v3, which is now platform-agnostic. It ships with a GitHub provider, but you can technically look for clankers beyond GitHub.
Matteo Gabriele
Matteo Gabriele
@matteogabriele.npmx.social
Always be yourself. Unless you can be Batman, in which case always be Batman. agentscan.tools • https://ai-policy.dev • npmx.dev https://github.com/matteogabriele
448 followers144 following2.5K posts
Matteo Gabriele reposted
🙋♂️ what if one client worked on every forge? 𝔣𝔬𝔯𝔤𝔢𝔰 is a new TypeScript library for issues, PRs, notifications, checks and webhooks across 12 forges ✨ forges.link
A few weeks ago, I shared the idea of analyzing contributors’ text for signs of automation. Now it's an app and a GitHub Action you can add to your anti-clanker Swiss Army knife. It's still an ongoing experiment, but it seems to capture the patterns in PR and issue descriptions reasonably well.
Since everyone is closing PR contributions, I am inviting everyone to spam my project with your best PR slop and impossible issues, so I can test new features against AI spam. I offer the repo; you bring the tokens. It's a win-win.
Another Sunday, another experiment. Interlinked tries to flag AI-written PR descriptions by looking for patterns found in PRs from known automated accounts and in descriptions written before AI. They don't give us the tools; we try to make a new one.
Another Sunday, another experiment. Interlinked tries to flag AI-written PR descriptions by looking for patterns found in PRs from known automated accounts and in descriptions written before AI. They don't give us the tools; we try to make a new one.
My AI contribution policy has been improved and is now part of ai-policy.dev, an open-source collection of AI policies. It's not a silver bullet. You can't enforce a policy programmatically, but it helps you and everyone contributing understand where your project stands.
Matteo Gabriele reposted
Today we're launching ai-policy.dev! A collaboration of me, @danielroe.dev and @matteogabriele.npmx.social - this provides a few human curated AI contribution policies you can choose from. Our hope is that this will help you better define your stance and what kind of contributions you accept 🙏
Matteo Gabriele reposted
If you are in Amsterdam this Thursday, come to the @vitest.dev x @npmx.dev meetup! guild.host/events/vites...
Matteo Gabriele reposted
I haven't post anything from @zurichjs.com because I was too busy enjoying Zurichjs. I had a blast meeting so so so many beautiful people ❤️
Matteo Gabriele reposted
AgentScan data, gathered hourly from 400 repos, clearly shows that for every PR you receive during the week-end, 1/3 to 1/2 will most likely be from an agent. Now using AgentScan helps OSS maintainers triage, to keep week-ends enjoyable.
GitHub doesn't show how many projects use your GitHub Action or GitHub App, so I wrote a custom script to retrieve that information for AgentScan. I didn't expect this many projects ❤️ I'm stoked on one end and sad on the other for the mess AI spam is causing. Stay positive 💪
Inspired by @jviide.iki.fi post about GitHub name typosquatting, I made a prototype of a possible action we could use in our workflows 👀
It would also be interesting to have a list of influential developers, like @43081j.com, with complex names that could be easily misinterpreted
Inspired by @jviide.iki.fi post about GitHub name typosquatting, I made a prototype of a possible action we could use in our workflows 👀
original post here bsky.app/profile/jvii...
It's, uh, less than ideal that I'm allowed to claim a GitHub username like this.
Matteo Gabriele reposted
Wrote a userscript that adds "AI" labels to users detected by agentscan.tools (by @matteogabriele.npmx.social)! Much needed hints to judge "contributions" these days. 👉 github.com/bluwy/github...
The Voight-Kampff test is back. If you want a package that handles GitHub requests and account activity analysis, you can use this one now. github.com/unveil-proje...
Matteo Gabriele reposted
i've written up some thoughts about AI in the open source world, and more importantly, why humanity is important. its a bit wordy and doomy but this is an important topic right now. there is still a lot of good in the OSS world at least ❤️