skills/paulnsorensen/easy-cheese/mold/Gen Agent Trust Hub

mold

Pass

Audited by Gen Agent Trust Hub on May 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill executes a bundled Python executable (mold.pyz) to perform path resolution and analyze specification files. A review of the source code within the executable confirms it only uses standard Python libraries and read-only git commands for project identification.
  • [SAFE]: The skill incorporates comprehensive safety gates, including the 'Two-key handshake' which mandates explicit user approval before any file writes, and 'Agent-introduced scope' checks to ensure no unauthorized features are added to specifications.
  • [SAFE]: No patterns of data exfiltration, prompt injection, or malicious remote code execution were detected. Environment variable usage is restricted to standard project and XDG path configuration.
  • [SAFE]: The skill's primary function is to manage local design artifacts and documentation, aligning with its stated purpose and the provided author context.
Audit Metadata
Risk Level
SAFE
Analyzed
May 29, 2026, 05:39 AM
Security Audit — agent-trust-hub — mold