async-interview-automation
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS] (MEDIUM): The skill requires connecting to an external, unverified MCP server at
https://rube.app/mcp. This server provides the tools and logic for the skill's operation. - [PROMPT_INJECTION] (HIGH): High risk of indirect prompt injection (Category 8). The skill processes untrusted data (interview responses/tasks) while possessing capabilities to execute actions. Ingestion points: External content fetched via
RUBE_MULTI_EXECUTE_TOOL. Boundary markers: None mentioned or enforced. Capability inventory: Execution of interview operations and remote workbench access viaRUBE_MULTI_EXECUTE_TOOLandRUBE_REMOTE_WORKBENCH. Sanitization: None. - [COMMAND_EXECUTION] (HIGH): The skill enables execution of actions through the Composio toolkit and a remote workbench. Furthermore, it relies on dynamically discovered 'execution plans' from the remote server, which could be manipulated to execute unauthorized commands.
Recommendations
- AI detected serious security threats
Audit Metadata