Skills
skills/s-hiraoku/vscode-sidebar-terminal/skill-creator/Gen Agent Trust Hub

skill-creator

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses subprocess calls to execute its own helper scripts and the Claude CLI for evaluation and packaging tasks.
  • eval-viewer/generate_review.py uses subprocess.run with lsof to manage the review server port.
  • scripts/run_eval.py uses subprocess.Popen to invoke the claude CLI for trigger testing.
  • SKILL.md instructs the agent to run various Python scripts for benchmark aggregation, packaging, and optimization loops.
  • [EXTERNAL_DOWNLOADS]: The skill references standard web dependencies for its HTML viewer (Google Fonts, SheetJS CDN).
  • [REMOTE_CODE_EXECUTION]: The skill utilizes subagents to execute test prompts and grading tasks, which is the intended functional design for its iterative improvement loop.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 07:32 AM