mobile-browser-workflow-generator
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection by processing external codebase data and web search results through sub-agents without sanitization.
- Ingestion points: Project codebase analyzed in Phase 2 and WebSearch results obtained in Phase 4.
- Boundary markers: Not present in the sub-agent task prompts.
- Capability inventory: Spawning sub-agents, executing web searches, and writing to the local file system.
- Sanitization: No evidence of input filtering or escaping is provided before processing the discovered content.
Audit Metadata