data-routing

Pass

Audited by Gen Agent Trust Hub on Nov 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill is designed to connect to and download data from a wide array of external financial services and APIs, including Yahoo Finance, OKX, Nobitex, Wallex, Tushare, Akshare, Baostock, and others.
  • [DYNAMIC_EXECUTION]: The instructions direct the agent to generate and execute Python code for materializing SDK results, performing custom calculations, and writing CSV artifacts using the managed_reader pattern and internal modules like src.market_data.fetch_market_data.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external providers through tools like get_stock_news, get_research_reports, and get_cn_announcements, which could contain malicious instructions.
  • Ingestion points: Data enters the context via multiple specialized tools (e.g., news and report fetching) described in SKILL.md.
  • Boundary markers: The skill mentions that errors and data are returned within an "envelope" structure.
  • Capability inventory: The system possesses capabilities for Python execution, file system writes (CSV), and shell execution via bash as noted in the guidelines.
  • Sanitization: No specific methods for sanitizing or filtering text-based news or report data are described.
  • [COMMAND_EXECUTION]: The skill guidelines reference the use of bash and background_run for executing acquisition and calculation tasks, indicating a shell command execution capability.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 21, 2026, 05:36 PM