orca-exposure-map
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes data from cloud resources to generate remediation scripts, creating a surface for indirect prompt injection.
- Ingestion points: Results from discovery_search and asset-related summaries in SKILL.md.
- Boundary markers: No explicit markers or instructions to ignore embedded commands are used for interpolated data.
- Capability inventory: File writing (e.g., lockdown scripts) and cloud discovery tool execution.
- Sanitization: No mentioned validation of cloud data before script generation.
- [COMMAND_EXECUTION]: The skill dynamically generates security remediation scripts in multiple formats based on discovered asset data.
Audit Metadata