servicenow
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Instructs the user to install the
@membranehq/clitool globally via npm, which is the official interface for the Membrane platform. - [COMMAND_EXECUTION]: Employs the
membraneCLI to authenticate, establish connections, and perform operations within ServiceNow, such as listing records or executing actions. - [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it retrieves and processes data from external ServiceNow sources.
- Ingestion points: ServiceNow API endpoints (Incidents, Tasks, Users) accessed via
membrane action runandmembrane requestin SKILL.md. - Boundary markers: No explicit delimiters or safety instructions are used to separate fetched data from agent instructions.
- Capability inventory: Uses the Membrane CLI to execute actions and network requests based on retrieved or provided data.
- Sanitization: No evidence of input sanitization or validation is present in the skill markdown.
Audit Metadata