skill-creator-advanced

Pass

Audited by Gen Agent Trust Hub on Apr 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The helper script scripts/infer_destination.py uses subprocess.run to execute git rev-parse commands. This is a legitimate operation used to identify the repository root for correct placement of new skills.
  • [DATA_EXFILTRATION]: The skill performs local discovery of common installation paths (e.g., ~/.claude/skills, ~/.agents/skills) to recommend destination roots. This data remains local and is used solely for the tool's intended path-aware logic.
  • [EXTERNAL_DOWNLOADS]: The metadata.json and various reference files point to repositories and documentation hosted by well-known technology providers such as Vercel Labs and agentskills.io. These are recognized as trusted external resources for the AI agent development ecosystem.
  • [SAFE]: All templates and instructions follow security best practices, such as recommending the use of environment variables for API keys and providing non-functional placeholders in code examples.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 16, 2026, 07:24 AM