explain-with-preview
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the execution of a Python script to process user-provided JSON files.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from JSON files that can contain HTML tags, creating a surface for indirect instructions.
- Ingestion points: Data from decide.json.
- Boundary markers: No explicit markers are defined to isolate data from instructions.
- Capability inventory: Local command execution and file writing.
- Sanitization: The skill instructions permit raw HTML tags like anchor tags, which could be used for malicious content.
Audit Metadata