generate-judge-contest
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes
bunx exercode-problemfor validating both problems and contest configurations. This is a standard and safe use of CLI tools for content verification within a development workflow. - [INDIRECT_PROMPT_INJECTION]: The skill generates contest data based on user-provided instructions and existing course context, which represents an attack surface for indirect prompt injection.
- Ingestion points: User-provided contest parameters and existing course/lecture materials.
- Boundary markers: Absent.
- Capability inventory: Includes file writing (YAML and problem directories) and shell command execution via
bunxfor validation. - Sanitization: The skill incorporates a validation step using the
exercode-problemCLI tool to ensure generated files meet the expected schema and constraints.
Audit Metadata