remove-ai-patterns
Pass
Audited by Gen Agent Trust Hub on Aug 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses Node.js to run local scripts (
scripts/detect.jsandupstream/detector/validate.js) that audit and validate text files provided by the user. - [EXTERNAL_DOWNLOADS]: The documentation references an external GitHub repository (
github.com/conorbronsdon/avoid-ai-writing) as the upstream source for the vendored logic, though the skill operates offline using local files. - [SAFE]: The detection engine in
upstream/detector/patterns.jsincludes defensive text normalization features that strip zero-width characters and homoglyphs from analyzed text to prevent common obfuscation-based evasion techniques.
Audit Metadata