User avatar
neobot_flag_trans Polyphony Cyberwitchery & Software asparklessparkles_pink @polyphonychat@amazonawaws.com
4w
`repo-slopscan` update; fighting ai code (+) repo-slopscan has been updated with a more accurate detections database! More agents get detected and the (by far) most common false-positive has been eliminated.

Repositories scanned with outdated detections will display a message and button like seen in the attached image, prompting you to re-scan this repository for more accurate results ablobcatwobwork
This note is quoting fed.amazonawaws.com/notes/alsi1chf1zq8accw
Screenshot. Text: The results on this page are outdated. The AI detection token definition files have been updated! Re-scan this repository for more accurate results. Button with text: Re-scan this repository
3
8
16
0
User avatar
🌸 lumi-nhac catbite no_ai antifa 🌱 @lumi@snug.moe
4w
`repo-slopscan` update; fighting ai code (+) @polyphonychat how does the scoring work?
1
0
0
0
User avatar
neobot_flag_trans Polyphony Cyberwitchery & Software asparklessparkles_pink @polyphonychat@amazonawaws.com
4w
`repo-slopscan` update; fighting ai code (+) @lumi Each detected signal is assigned a signal type (for AI authorship, AI co-authorship, etc.), which in turn is assigned penalty weight value. For each occurrence of a signal, that penalty is deducted from the perfect (100) score. This gets repeated for the entire commit history, or up to 5000 commits.
0
0
0
0