agent-startup-advisor
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interpolates user-supplied input directly into its execution flow via the
$ARGUMENTSplaceholder. - Ingestion points: Untrusted data enters the agent context through the
$ARGUMENTSvariable at the end ofSKILL.md. - Boundary markers: There are no protective delimiters or instructions to ignore embedded commands surrounding the interpolated input.
- Capability inventory: The skill is configured for a
general-purposeagent, which typically possesses broad file and tool access capabilities. - Sanitization: No sanitization, escaping, or schema validation is applied to the input before it is processed by the agent.
Audit Metadata