team-level

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: No security issues or malicious patterns were detected. The skill implements mandatory human-in-the-loop approval via AskUserQuestion for all major decision points.
  • [PROMPT_INJECTION]: The skill ingests content from local game design documents, which creates a surface for potential indirect prompt injection.
  • Ingestion points: Reads documents from design/gdd/, design/levels/, and design/narrative/ to provide context to sub-agents.
  • Boundary markers: No delimiters are used to separate ingested content from agent instructions.
  • Capability inventory: The skill allows usage of Write, Edit, Bash, and Task tools.
  • Sanitization: All sub-agent outputs and step transitions require explicit user approval via AskUserQuestion.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 06:29 PM