priority-judge
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the user and outputs it to a persistent local file, creating a vulnerability surface for potential indirect prompt injection.
- Ingestion points: User-provided to-do items and task descriptions gathered during the 'Collection' and '詢問' phases (SKILL.md, Steps 1 and 2).
- Boundary markers: The skill does not provide specific delimiters or instructions for the agent to ignore or isolate potential instructions embedded within the user's task text.
- Capability inventory: The skill instructs the agent to perform file-writing operations to save markdown priority lists in the '计划探索/' directory (SKILL.md, Step 5).
- Sanitization: The skill lacks instructions for sanitizing, escaping, or validating user input before it is recorded in the final markdown document.
Audit Metadata