re-amaze-automation

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and fetches tool definitions from the external endpoint https://rube.app/mcp and toolkit documentation at https://composio.dev/toolkits/re_amaze.\n- [REMOTE_CODE_EXECUTION]: Instructions direct the user to connect to a remote Model Context Protocol (MCP) server at https://rube.app/mcp, enabling the agent to execute tools hosted on the remote Rube platform.\n- [COMMAND_EXECUTION]: The workflow employs RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH to perform actions on the Re Amaze service based on runtime-discovered schemas.\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it processes untrusted data from the Re Amaze platform.\n
  • Ingestion points: External data from Re Amaze ticket and message operations (SKILL.md).\n
  • Boundary markers: The skill lacks explicit delimiters or instructions to the agent to ignore potentially malicious commands embedded in the ingested data.\n
  • Capability inventory: Automated execution capabilities via RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH (SKILL.md).\n
  • Sanitization: No sanitization or validation procedures for external content are defined in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 05:42 AM