workable-automation
Warn
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
- EXTERNAL_DOWNLOADS (MEDIUM): The skill instructions direct the user to add an external, non-trusted MCP server endpoint (
https://rube.app/mcp). This server provides the tool definitions and logic used by the agent. - REMOTE_CODE_EXECUTION (MEDIUM): The skill references the
RUBE_REMOTE_WORKBENCHtool. Such tools typically allow for remote code execution (RCE) within a controlled environment, but the skill lacks visibility into the security boundaries of this remote environment. - DATA_EXFILTRATION (LOW): Authentication for the Workable toolkit is managed via
RUBE_MANAGE_CONNECTIONS. While no secrets are hardcoded in the skill, the process involves sending authentication flows through the externalrube.appinfrastructure. - Indirect Prompt Injection (LOW): The skill relies on fetching dynamic tool schemas and "recommended execution plans" via
RUBE_SEARCH_TOOLS. This creates an ingestion point where an attacker-controlled MCP server could influence the agent's behavior by providing malicious instructions within the tool metadata.
Audit Metadata