tradecraft-bugbounty-platforms
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze external, untrusted data from bug bounty program policies and scope pages.
- Ingestion points: External policy pages, scope documentation, and program URLs from platforms such as HackerOne, Bugcrowd, and Immunefi.
- Boundary markers: The instructions lack specific guidance for the agent to use delimiters or warnings to ignore potentially malicious instructions embedded within the external policy documents.
- Capability inventory: While this specific reference skill does not invoke dangerous tools, it is designed to be active during bug bounty workflows where the agent may have access to network tools and file systems.
- Sanitization: No sanitization or escaping mechanisms are defined for the content retrieved from external sources.
Audit Metadata