skills/tmeister/skills/vault-ideas/Gen Agent Trust Hub

vault-ideas

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized data exfiltration attempts were detected. The skill consists of instructional content without executable code.
  • [COMMAND_EXECUTION]: The skill instructions guide the agent to perform local file system operations (reading and writing markdown files) strictly within a user-confirmed directory.
  • [PROMPT_INJECTION]: While the skill scans user-authored notes which represents an indirect prompt injection surface, the risk is negligible as the skill lacks network permissions or administrative shell access to facilitate exploitation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 12:22 AM