story-short-write
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill contains instructions for the agent to execute a shell command using Python to calculate word counts for local story files. The command
python -c "from pathlib import Path; print(len(Path('file_path').read_text(encoding='utf-8')))"is used purely for document length verification. - [REMOTE_CODE_EXECUTION]: The skill utilizes local Node.js scripts (
scripts/normalize-punctuation.jsandscripts/check-ai-patterns.js) to automate punctuation normalization and AI-pattern detection in the generated text. These scripts operate locally on the project files. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates a multi-step writing process that involves reading analyzed data from previous steps. It includes specific directives to prevent the agent from including internal analysis markers or self-correction logs in the final user-facing output, mitigating the risk of instructions leaking into the narrative.
Audit Metadata