authority
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of Markdown instructions and JSON test cases. There are no scripts (Python, JavaScript, Shell) or binary executables associated with this skill.
- [SAFE]: The content is purely instructional, providing a methodology for agents to help users recognize and evaluate authority-based influence as described in Robert Cialdini's book 'Influence'. It does not request any special permissions or tool access.
- [INDIRECT_PROMPT_INJECTION]: While the skill defines 'Language signals' to trigger its logic based on user input, it does not possess any capabilities (such as file system access or network requests) that could be exploited via injection. The risk is limited to the conversational context.
- [METADATA_POISONING]: The metadata fields (name, description, tags) are consistent with the instructional content and do not contain hidden commands or deceptive instructions.
Audit Metadata