pilvio-automation
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches tool schemas and execution plans from Composio's official MCP infrastructure at rube.app.
- [COMMAND_EXECUTION]: Instructs the agent to execute tools on the Pilvio platform using RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH.
- [PROMPT_INJECTION]: Utilizes dynamic discovery which ingests execution plans and tool schemas from a remote server into the agent's context. This architecture creates a surface for indirect prompt injection. Ingestion points: RUBE_SEARCH_TOOLS output (SKILL.md). Boundary markers: Absent. Capability inventory: RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH (SKILL.md). Sanitization: Not specified.
Audit Metadata