soak-test
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_CONTEXT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from multiple local files which could contain malicious instructions. Ingestion points: Reads project.yaml, design/gdd/game-concept.md, and files in production/qa/ (SKILL.md). Boundary markers: No explicit delimiters or instructions to ignore embedded content are present. Capability inventory: The skill can write files and execute a restricted project script via Bash. Sanitization: No input validation or escaping is applied to ingested data.
- [DYNAMIC_CONTEXT_INJECTION]: The skill executes a project-local helper script at load time to resolve YAML configurations via the exclamation-backtick syntax. This is identified as a legitimate use of project-specific tooling and is considered safe.
Audit Metadata