council
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS] (LOW): The skill recommends installing official CLI tools from trusted organizations including Anthropic, OpenAI, and Google. Following the trusted source rule, these dependencies are categorized as low risk.
- [COMMAND_EXECUTION] (LOW): Example bash scripts and patterns use variable interpolation (e.g., $CODE, $QUESTION) directly within command strings. If an agent executes these with unsanitized user input containing shell metacharacters, it could lead to unintended local command execution.
Audit Metadata