skills/ziyueyijun/skills/powerbuilder/Gen Agent Trust Hub

powerbuilder

Pass

Audited by Gen Agent Trust Hub on Nov 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, prompt injections, or data exfiltration attempts were detected in the instructions or reference materials.
  • [COMMAND_EXECUTION]: Includes a Python script (scripts/pb_check.py) designed for static analysis of PowerScript source files. The script performs read-only regex-based checks for common anti-patterns (e.g., NULL comparisons, missing transaction commits) and does not execute the analyzed code or perform any destructive operations.
  • [EXTERNAL_DOWNLOADS]: Mentions official Sybase/SAP licensing and support URLs, as well as standard third-party dependencies for PDF generation (Ghostscript, Apache FOP). These are legitimate references to well-known technology services and do not pose a security risk.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided PowerBuilder code for analysis and generation. While this constitutes an attack surface, the skill implements a structured workflow ("Six-step workflow") and references local "hard rules" (coding-rules.md) to ground the agent's behavior and minimize the risk of accidental obedience to embedded instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 18, 2026, 01:30 AM