seyedehsanhadi/sloptrim ? reverse-engineered prompt

Reverse engineered prompt

Build me a local writing checker for AI sounding prose, something I can run inside Claude Code and Cursor that scores saved text files and points out the awkward spans I should fix.

I want it to work only on prose, not code, and stay completely on my machine, with no network and no model calls. It should be able to inspect common writing files, including markdown and docx, give each file a 0 to 100 style score, name the patterns it noticed, and tell me when the writing feels clean, mixed, or too obviously machine written. If possible, make it easy to install as a plugin or command, with a simple doctor check and a default mode that watches saves and nudges me when the score is too high.

Use only the Python standard library if you can, keep it lightweight, and if you need current docs or exact agent install steps, look them up online first.

Are you gonna build this?

make sure you review the code using coderabbit

Try freeSponsored — opens CodeRabbit in a new tab