The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local Node.js script (scripts/detect.js) to perform text analysis. The execution is localized to the user's environment and is used to provide deterministic analysis results for prose. Evidence: printf '%s' "$TEXT" | node scripts/detect.js --context general and node scripts/detect.js --file path/to/draft.md in SKILL.md.
[OBFUSCATION]: The scripts/patterns.js file contains logic to detect and normalize common obfuscation techniques used to bypass AI detectors, such as zero-width characters and homoglyphs. This is a defensive security feature within the skill's logic, intended to improve detection accuracy by stripping invisible characters and mapping Cyrillic/Greek lookalikes back to Latin characters before analysis. Evidence: normalizeText function in scripts/patterns.js including CYRILLIC_LOOKALIKES and ZERO_WIDTH_GLOBAL_RE.

ai-writing-detector