story-review
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze user-provided prose drafts, which inherently creates a surface for indirect prompt injection if the processed content contains malicious instructions.\n
- Ingestion points: Markdown files containing prose drafts are read by the
analyze.pyscript and reviewed by the agent as part of the story review process.\n - Boundary markers: The skill provides detailed methodological resources and checklists for different review levels (editorial, copyedit, etc.) that help guide the agent's focus and isolate the draft content from its operational instructions.\n
- Capability inventory: The skill utilizes a local Python script (
resources/prose-critique/analyze.py) to perform mechanical prose analysis. It does not request network access or high-privilege system operations.\n - Sanitization: The analysis script includes functions to strip markdown formatting and YAML frontmatter from the text before calculating metrics, providing a basic level of data isolation.
Audit Metadata