skills/hoetaek/leaf/leaf-work/Gen Agent Trust Hub

leaf-work

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a local command-line interface named leaf for initializing projects, managing state, and performing diagnostics. These operations are restricted to a local workspace directory (.leaf/).
  • [EXTERNAL_DOWNLOADS]: The documentation references external best-practice guides from well-known technology organizations (GitHub, Google, SmartBear) for informational purposes. It also describes a workflow for capturing rendered HTML from user-specified web pages for UI design tasks, which is an intended core function of the skill.
  • [SAFE]: No evidence of prompt injection, obfuscation, unauthorized data exfiltration, or hardcoded credentials was found across the analyzed files. The instructions emphasize strict adherence to a structured 'gate' system with human-in-the-loop approval points.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 01:44 PM
Security Audit — agent-trust-hub — leaf-work