The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a Python script (scripts/check_output.py) to perform static analysis on text files. This script is invoked via python3 to identify and optionally fix stylistic issues. This is the primary intended function of the skill and operates on local text data.
[EXTERNAL_DOWNLOADS]: The skill references external research data from github.com/louisabraham/load-bearing and github.com/berenslab/llm-excess-vocab. These are used as authoritative sources for identifying 'AI slop' vocabulary. These references are informative and do not involve the execution of untrusted remote code.
[FALSE_POSITIVE_ANALYSIS]: Automated scanners flagged references/refresh-vocabulary.md and references/html-report.md as malicious. However, manual review shows these files contain documentation and regex patterns for detecting AI-generated text. The 'malicious' classification likely stems from the inclusion of 'jailbreak-like' or 'injection-like' phrases used as examples of what the skill should filter out. Similarly, the URL louisabraham.github.io/load-bearing is a legitimate research project tracking LLM vocabulary trends.
[SAFE_PRACTICES]: The skill explicitly instructs the agent to preserve code blocks, citations, and quoted speech during rewrites, preventing accidental corruption of technical data. It also uses standard Python libraries and follows best practices for local script execution.

rewrite-slop