mailcheck-automation

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • External Downloads (HIGH): The skill instructs the user to add an untrusted third-party MCP endpoint https://rube.app/mcp to their configuration. This source is not on the trusted provider list, making its logic unverifiable.
  • Remote Code Execution (HIGH): Through the MCP server, the agent executes logic provided by the remote endpoint. Tools such as RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH provide the remote service with direct execution capabilities over the agent's workflow.
  • Indirect Prompt Injection (HIGH): The skill is highly vulnerable to malicious data processed via the Mailcheck toolkit. Ingestion points: External data from Mailcheck tasks and operations. Boundary markers: No delimiters or isolation instructions are present to separate data from system instructions. Capability inventory: High-impact capabilities including RUBE_MULTI_EXECUTE_TOOL and remote workbench execution. Sanitization: No sanitization or validation of external input is implemented.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 12:47 PM