SlopOne research preview: tell us what you find

[fhwang]

SlopOne is a free research preview in the Qlty CLI. It scores the maintainability of every file in your repository by combining Qlty's static analysis with TypeSafe's Jev decision model. It shows how those scores change over time, and it turns the files you pick into a refactoring prompt for your coding agent.

Try it

curl https://qlty.sh | bash
export TYPESAFE_API_KEY=your-key
qlty slop-one

Using Vercel AI Gateway or OpenRouter instead? Set AI_GATEWAY_API_KEY or OPENROUTER_API_KEY, and add --provider vercel or --provider openrouter.

What we'd like to hear

Reply with whatever you find. These help us most:

• Where a score looks wrong. A file SlopOne scored well that you'd call hard to maintain, or the reverse. Tell us what the file does and why you disagree, and link to it if the repository is public.
• Where a score looks right in a way your current tools miss.
• How the refactor went. Which coding agent you used, and whether the prompt led to a better change.
• Your setup. Languages, rough repository size, which provider you used, and how long the first run took.

Please don't paste proprietary code here. Describe it, or link to a public repository.

Found a bug? Open an issue: https://github.com/qltysh/qlty/issues/new

Want to talk it through live? Join us on Discord: https://qlty.sh/discord

We read every reply. We'll publish the methodology and our measurements soon.