godial-automation
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill fetches tool schemas from a remote MCP server via 'RUBE_SEARCH_TOOLS', creating an ingestion point for potentially untrusted instructions that could influence the agent's execution of subsequent tools.
- Ingestion points: 'RUBE_SEARCH_TOOLS' response (SKILL.md);
- Boundary markers: Absent;
- Capability inventory: 'RUBE_MULTI_EXECUTE_TOOL' and 'RUBE_REMOTE_WORKBENCH' (SKILL.md);
- Sanitization: Absent.
Audit Metadata