buzz
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its method of passing user data to specialized tools.
- Ingestion points: Untrusted data enters the agent context through the
{{args}}variable inSKILL.md. - Boundary markers: The skill lacks any delimiters (such as XML tags or triple backticks) to isolate the user-provided
{{args}}from the internal instructions passed to sub-skills. - Capability inventory: The skill environment permits powerful tools including
Bash,WebFetch, andWebSearch, which could be targeted by a malicious payload embedded in the user input if it successfully overrides the agent's logic. - Sanitization: There is no evidence of validation or sanitization of the user input before it is interpolated into the command to invoke sub-skills.
Audit Metadata