ralph
Warn
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: MEDIUMDATA_EXFILTRATION
Full Analysis
- [DATA_EXFILTRATION]: The skill contains a directory traversal string
../../../.agents/workflows/ralph.md. This is a suspicious pattern used to escape the skill's sandbox directory in an attempt to read internal agent workflows, system prompts, or configuration files stored in the hidden.agentsdirectory.
Audit Metadata