epic-games-automation

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection (Category 8).
  • Ingestion points: Processes external data from tool responses provided by RUBE_SEARCH_TOOLS and RUBE_MULTI_EXECUTE_TOOL as defined in SKILL.md.
  • Boundary markers: The instructions do not define delimiters or specific ignore guidelines for the agent when processing content from these external tool outputs.
  • Capability inventory: The agent has access to powerful execution capabilities through the RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH commands.
  • Sanitization: No evidence of validation or sanitization is provided for data received from tool responses before it is used to influence the agent's workflow.
  • [SAFE]: The skill integrates with established services from Composio (composio.dev and rube.app). These resources are used for their primary intended purpose in a developer-focused automation context.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 03:38 AM