ustva-simulation
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, command injections, or exfiltration vectors were detected.
- [INDIRECT_PROMPT_INJECTION]: The skill includes instructions to process external data such as XML exports and invoice lists, representing a potential indirect prompt injection surface.
- Ingestion points: Processes XML exports and user-provided financial data (referenced in SKILL.md).
- Boundary markers: No specific delimiters or boundary markers are defined for the processed data.
- Capability inventory: The skill instructs the agent to generate template files in the
assets/templates/directory but does not request sensitive system access or network operations. - Sanitization: The skill includes a mandatory triage step requiring the agent to ensure all receipts are correctly anonymized for data protection purposes before simulation.
- [COMMAND_EXECUTION]: The skill does not contain any executable code or shell commands. It operates entirely through natural language instructions for the agent's behavior.
- [DATA_EXFILTRATION]: While the skill handles sensitive tax-related identifiers (Steuernummer) and financial data, it provides explicit warnings against real data submission and lacks any network capabilities to exfiltrate the processed information.
Audit Metadata