revise
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes user-provided markdown files, which constitutes an untrusted data surface.
- Ingestion points: The full article content is read into the agent context in the 'Analyze Structure' phase.
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when processing the article text.
- Capability inventory: The skill possesses file-write permissions for structural editing and can invoke internal tools such as '/prose-fix'.
- Sanitization: There is no evidence of content sanitization to prevent the agent from potentially obeying instructions embedded within the processed drafts, though the requirement for user confirmation before reordering provides a manual safety check.
Audit Metadata