first-ask
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to gather information through human input and web research, creating an attack surface where untrusted data could influence agent behavior.\n
- Ingestion points: Data enters the context via the
joyride_request_human_inputtool and web research/project exploration tools inSKILL.md.\n - Boundary markers: There are no explicit boundaries or instructions to ignore embedded prompts within the gathered information.\n
- Capability inventory: The skill leverages available system tools to perform tasks based on the input.\n
- Sanitization: No data sanitization or input validation mechanisms are defined within the instructions.
Audit Metadata