skills/clawic/skills/Folders/Gen Agent Trust Hub

Folders

Pass

Audited by Gen Agent Trust Hub on May 31, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides guidelines for local folder indexing and management. It includes explicit instructions for path canonicalization and rejecting access to sensitive system directories (e.g., /etc, /var, C:\Windows).
  • [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes untrusted directory names from the file system.
  • Ingestion points: Local directory traversal during the discovery process described in SKILL.md.
  • Boundary markers: No specific delimiters or markers are defined for directory metadata.
  • Capability inventory: File system traversal and move-to-trash operations.
  • Sanitization: The instructions explicitly mandate path canonicalization and system path blacklisting to prevent path traversal or sensitive file access.
  • [DATA_EXPOSURE]: The tool maintains a local index at ~/.config/folder-index.json. No external network requests or sensitive data exfiltration patterns were identified.
Audit Metadata
Risk Level
SAFE
Analyzed
May 31, 2026, 09:21 PM
Security Audit — agent-trust-hub — Folders