cro-research
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from external CSV files (e.g., GA4 exports).
- Ingestion points: The
scripts/cro_research.pyscript reads data from paths provided via--pages,--segments, or--funnelarguments. - Boundary markers: Data is processed and rendered into structured markdown tables and summaries for the agent's context.
- Capability inventory: The script's capabilities are strictly limited to mathematical analysis and standard output. It contains no network operations, file-writing functions, or subprocess execution capabilities.
- Sanitization: The script performs data validation, converting counts to numeric types and truncating/escaping labels (e.g., escaping pipe characters for markdown compatibility) before the agent processes the results.
Audit Metadata