cpo-review
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes local Python scripts to perform RICE scoring, PMF scoring, and portfolio analysis. The scripts are located at the following workspace paths:
product-team/skills/product-manager-toolkit/scripts/rice_prioritizer.py,../../../c-level-advisor/skills/cpo-advisor/scripts/pmf_scorer.py, and../../../c-level-advisor/skills/cpo-advisor/scripts/portfolio_analyzer.py. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data via the
<plan>argument, creating a potential vector for malicious instructions to be processed by the agent. - Ingestion points: The
<plan>argument within the/cs:cpo-reviewcommand inSKILL.md. - Boundary markers: Absent. The instructions do not specify delimiters or safety warnings to ignore instructions embedded within the plan.
- Capability inventory: The skill can execute local Python scripts via the shell, which could be exploited if the plan content influences command arguments or script execution flow.
- Sanitization: Absent. No evidence of input validation, filtering, or escaping was found in the skill definitions.
Audit Metadata