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