research-brand
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external websites and search results, which provides a surface for indirect prompt injection.
- Ingestion points: The skill crawls the user-provided URL and specific sub-pages (homepage, /about, /pricing, etc.), and extracts data from search results on platforms like Crunchbase, LinkedIn, and app stores as specified in
SKILL.md. - Boundary markers: There are no explicit instructions in the prompt to ignore instructions embedded within the crawled web content or to use specific delimiters for the external data.
- Capability inventory: The skill utilizes file-writing capabilities to save the
brand_dna.mdreport to the local file system. - Sanitization: The instructions do not define a process for sanitizing or validating the text extracted from external sources before it is processed by the agent.
Audit Metadata