game-development
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to research comparable games by watching gameplay, inspecting playable demos, and reading engine documentation. This process creates a vulnerability surface where instructions embedded within these external, untrusted materials could influence the agent's actions.
- Ingestion points: Data enters the agent's context through external game research, video analysis, and documentation inspection as described in
SKILL.md. - Boundary markers: The instructions lack specific boundary markers or warnings to disregard instructions found within external research materials.
- Capability inventory: The agent is authorized to modify game code, use browser tools for research, and execute/playtest the software it develops.
- Sanitization: There is no mention of sanitizing or validating the integrity of content retrieved from external research sources.
Audit Metadata