cn-finance-data

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses a surface for potential indirect prompt injection because it retrieves untrusted data from external sources such as news, corporate announcements, and investor Q&A APIs.
  • Ingestion points: pro.news, pro.anns_d, and pro.irm_qa_sz/sh methods fetch text from the web to be processed by the agent.
  • Boundary markers: Absent; there are no specific instructions to the agent on how to handle potential embedded instructions in the fetched data.
  • Capability inventory: The skill allows Python code execution via the Bash(python:*) tool.
  • Sanitization: Not explicitly implemented in the provided skill instructions.
  • [EXTERNAL_DOWNLOADS]: The skill documentation correctly guides the user to install well-known and widely used Python packages (tushare, pandas, openpyxl) from the official Python Package Index (PyPI).
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 12:27 PM
Security Audit — agent-trust-hub — cn-finance-data