threejs-game-workflow-control

Pass

Audited by Gen Agent Trust Hub on Oct 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes system commands such as git (for state tracking and diffing) and node (for running unit tests) using Node.js child_process modules. These actions are triggered as part of the implementation unit lifecycle and are restricted to paths and commands defined within the skill's internal configuration files.
  • [EXTERNAL_DOWNLOADS]: The skill depends on the sharp package (v0.35.3) for image manipulation tasks. This is a widely used and trusted library in the Node.js ecosystem.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes structured data from various JSON files (e.g., work-item.json, execution-state.json) which contain user-supplied descriptions and metadata. Potential risks are mitigated by enforcing strict JSON Schema validation and using dedicated contract verification scripts that isolate semantic content from executable logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 23, 2026, 02:05 PM