build-competitive-brief
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- Dynamic HTML Verification: The workflow supports generating self-contained HTML files and optionally verifying their appearance using a browser-based renderer. This involves the dynamic execution of generated content as part of the report validation process.\n- Ingestion of External Data: The skill processes information from various sources including CRM systems, internal messaging, and public research. This activity creates a potential surface for indirect prompt injection, which is a common security consideration when agents aggregate and summarize content from external environments.
Audit Metadata