shorten-rest-automation
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (HIGH): The skill instructs the agent to ingest and follow 'recommended execution plans' and schemas from
RUBE_SEARCH_TOOLS(provided byhttps://rube.app/mcp). Evidence Chain: 1. Ingestion points:RUBE_SEARCH_TOOLSoutput. 2. Boundary markers: Absent. 3. Capability inventory:RUBE_MULTI_EXECUTE_TOOLandRUBE_REMOTE_WORKBENCH. 4. Sanitization: Absent. - [External Downloads / Unverifiable Dependencies] (MEDIUM): The skill requires the user to add an external MCP server at
https://rube.app/mcp, which is not a trusted source per [TRUST-SCOPE-RULE]. - [Dynamic Execution] (MEDIUM): The use of
RUBE_MULTI_EXECUTE_TOOLandRUBE_REMOTE_WORKBENCH(SKILL.md) allows for the execution of commands and logic defined by the external server at runtime, which may include code execution or privileged operations.
Recommendations
- AI detected serious security threats
Audit Metadata