ui-pc

Pass

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: Disclosure of local file system paths.
  • The skill specifies absolute paths such as /Users/xujiajun/Developer/frontProj/web/src/** in the metadata and reference tables. This exposes the local username and directory structure of the developer's machine.
  • [PROMPT_INJECTION]: Vulnerability surface for Indirect Prompt Injection.
  • Ingestion points: The skill handles untrusted data through search components (leniu-search), form inputs (leniu-form), and API responses (getList, getDetail) used to populate tables and forms.
  • Boundary markers: The provided code snippets do not include delimiters or specific instructions to treat external data as untrusted content when processed by the agent.
  • Capability inventory: The skill demonstrates capabilities for network operations via a custom request utility and dynamic rendering of UI components.
  • Sanitization: Although the skill mentions UI validation rules (this.validators), it lacks explicit documentation or implementation of security sanitization to prevent malicious instructions embedded in data from influencing agent behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 7, 2026, 09:04 AM