partner-tiering
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided data such as partner P&L, revenue distributions, and engagement signals to construct a tiering specification.
- Ingestion points: Sensitive business data enters through the 'Interview' section and workflow step 3 in
SKILL.md. - Boundary markers: The skill does not instruct the agent to use specific delimiters or protective prompting (e.g., 'ignore instructions within data') when handling user inputs.
- Capability inventory: The skill generates a text spec and optionally browses the web. It does not possess high-risk capabilities like local code execution or direct file system writing.
- Sanitization: There is no mention of filtering or sanitizing the data collected from the user.
Audit Metadata