shortpixel-automation
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill instructs the agent to fetch tool schemas and execution plans from an external service (rube.app) and execute them dynamically. This creates a surface for indirect prompt injection if the external source is compromised.
- Ingestion points: Outputs from the
RUBE_SEARCH_TOOLSfunction as described in Step 1 of the workflow. - Boundary markers: The instructions do not define delimiters or specific constraints to ignore instructions embedded within the retrieved schemas.
- Capability inventory: The skill utilizes
RUBE_MULTI_EXECUTE_TOOLandRUBE_REMOTE_WORKBENCH, which are high-capability tools that act on the ingested data. - Sanitization: No sanitization or validation logic is defined for the content returned by the MCP server before it is used to generate tool arguments.
- [External Downloads] (SAFE): The skill references an external MCP endpoint
https://rube.app/mcp. This is used for service configuration and tool discovery rather than direct remote code execution or package installation. - [No Code] (SAFE): No scripts or binary files are included with this skill; it consists entirely of configuration and instructional markdown.
Audit Metadata