rabbithole-doc-exploration
Pass
Audited by Gen Agent Trust Hub on Jul 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for document exploration and includes proactive safety instructions. Specifically, it directs the agent to avoid pasting secrets, private profiles, or sensitive local files into the interactive canvas unless explicitly approved by the user.
- [SAFE]: Data access is limited to standard repository documentation and configuration files (e.g., AGENTS.md, RUNBOOK.md, .set-plan files) necessary for the skill's primary purpose. There are no indications of unauthorized data exfiltration or access to sensitive system paths.
- [SAFE]: The skill does not perform any external network operations, download remote code, or execute arbitrary system commands beyond the defined model context protocol (MCP) tool usage.
Audit Metadata