pitch-the-fix
Pass
Audited by Gen Agent Trust Hub on Nov 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to run a Python script for brand context loading. This is a legitimate configuration step using a local vendor-provided script.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from business websites and Google reviews which could contain malicious instructions.
- Ingestion points: Scraping of third-party About and Team pages and reading Google review text.
- Boundary markers: No delimiters or isolation instructions are provided for the scraped content.
- Capability inventory: File system writing, local command execution, and network scraping tools.
- Sanitization: No explicit sanitization of external content is mentioned.
Audit Metadata