startup-business-analyst-market-opportunity
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill instructions direct the agent to use WebSearch and WebFetch tools to gather data from external domains (Step 4 of SKILL.md). These operations target non-whitelisted domains.
- [PROMPT_INJECTION]: The skill facilitates an indirect prompt injection surface by retrieving and processing external content from the web without explicit sanitization or boundary markers.
- Ingestion points: WebSearch and WebFetch are used to retrieve industry reports, public filings, and other external data.
- Boundary markers: The skill lacks instructions to wrap or delimit external content to prevent it from overriding agent instructions.
- Capability inventory: The skill has access to tools including Bash, Write, and Edit, which provide a significant capability surface if an indirect injection were successful.
- Sanitization: There are no documented steps for filtering or validating the content retrieved from external sources before it is processed by the agent.
Audit Metadata