passslot-automation
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- Prompt Injection (HIGH): The skill requires the agent to fetch recommended execution plans and known pitfalls from the remote server rube.app. This untrusted data directly influences agent behavior. Evidence: 1. Ingestion point: RUBE_SEARCH_TOOLS (SKILL.md). 2. Boundary markers: None. 3. Capability inventory: RUBE_MULTI_EXECUTE_TOOL, RUBE_REMOTE_WORKBENCH (SKILL.md). 4. Sanitization: None.
- External Downloads (MEDIUM): The setup instructions mandate connecting to an untrusted external endpoint (https://rube.app/mcp) to provide the skill's core logic and tool schemas.
- Command Execution (MEDIUM): The skill implements tools for multi-tool execution and remote workbench access, providing a high-impact target for the identified injection surface.
Recommendations
- AI detected serious security threats
Audit Metadata