worldview-game-wounds-infection-and-treatment
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions are focused on architectural design for fictional game systems. It includes rigorous ethical guidelines to avoid dehumanizing narratives and strict prohibitions against providing clinical medical guidance. All processes revolve around generating documentation and configuration files without executing arbitrary commands or accessing sensitive data.- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied game briefs to generate mechanics, which represents a standard ingestion surface. However, it incorporates several design-level mitigations.
- Ingestion points: User-provided world, project, and route briefs following the slash command as described in
SKILL.md. - Boundary markers: The skill enforces a sequence of five 'Locks' (Fiction, Condition Model, Clock Policy, Treatment Route, and Persistence) that prevents input from redefining core safety or logic boundaries once established.
- Capability inventory: The skill's primary capabilities are writing Markdown and YAML files to specific project directories; it does not perform network operations or dynamic code execution on user input.
- Sanitization: The agent is instructed to strictly separate project facts, runtime observations, fictional proposals, and reviewed claims to prevent data confusion.
Audit Metadata