data-routing

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill consists entirely of markdown-based configuration and guidance. It contains no scripts, binary files, or command execution patterns, eliminating the risk of direct malicious code execution.
  • [DATA_EXPOSURE]: The skill correctly manages authentication by referencing environment variable names (e.g., TUSHARE_TOKEN, FINNHUB_API_KEY) rather than hardcoding sensitive credentials or API keys. This follows standard secure practices for secret management.
  • [INDIRECT_PROMPT_INJECTION]: The skill documents ingestion points for external financial news, filings, and report data. While this creates a potential surface for indirect prompt injection, it is the primary intended function of the skill for research purposes and does not pose an inherent risk in this routing context.
  • [SAFE]: The skill includes explicit instructions for data verification, requiring the cross-checking of figures across multiple independent sources and flagging mismatches, which enhances the reliability and security of the agent's output.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 02:30 PM
Security Audit — agent-trust-hub — data-routing