stepps-diagnostic
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user input (descriptions of marketing campaigns, products, or content) to perform a virality analysis.
- Ingestion points: User-provided data is collected in the 'Execution' steps of
SKILL.mdto be analyzed against the STEPPS framework. - Boundary markers: The instructions do not specify any delimiters (e.g., XML tags or triple quotes) or 'ignore' instructions to prevent the agent from obeying commands embedded within the user's content.
- Capability inventory: The skill possesses no dangerous capabilities such as file system writes, network operations, or subprocess execution.
- Sanitization: No methods for sanitizing, escaping, or validating the user-provided data are implemented.
Audit Metadata