tisane-automation

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): The skill requires the user to add an external MCP server endpoint (https://rube.app/mcp). While this is a documented requirement for using Rube MCP, it creates a dependency on an external third-party service not included in the trusted source list.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The skill relies on dynamic tool discovery via RUBE_SEARCH_TOOLS to determine execution plans.
  • Ingestion points: Data returned from the RUBE_SEARCH_TOOLS call (schemas, execution plans).
  • Boundary markers: Absent; the agent is instructed to use exact field names and types from the search results without explicit verification.
  • Capability inventory: Includes RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH, which allow for complex tool execution and remote operations.
  • Sanitization: None mentioned; the agent is expected to follow schemas provided by the remote discovery service.
  • [DYNAMIC_EXECUTION] (SAFE): While the skill mentions RUBE_REMOTE_WORKBENCH and run_composio_tool(), these are standard components of the Composio ecosystem and are used as intended for tool execution rather than arbitrary code generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:40 PM