build-game-ui-ux
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface when rendering networked or user-supplied strings (player names, chat logs, floating combat text).
- Ingestion points: Networked or user-supplied strings mentioned in the ASI05 RCE Guard section of SKILL.md.
- Boundary markers: Instructions specify using safe text nodes rather than raw rich text.
- Capability inventory: The skill uses standard file manipulation and command execution tools (write_file, execute_command) for development tasks.
- Sanitization: Explicitly mandates escaping markup and using safe rendering nodes to prevent injection attacks.
- [COMMAND_EXECUTION]: The skill configuration allows for command execution and build tools (run_build, execute_command). These are restricted to standard development workflows as defined in the YAML frontmatter and do not show signs of arbitrary or malicious execution patterns.
- [DATA_EXPOSURE]: The skill emphasizes validating and resetting cached HUD state on scene transitions (ASI06) to prevent stale or tampered data from influencing UI state, which is a defensive measure against local state corruption.
Audit Metadata