market-overview-dashboard

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill workflow in references/data-queries.md requires the agent to execute shell commands to activate a specific Python virtual environment and run the views_runner.py script located in a sibling directory (../findata-toolkit-cn/).
  • [EXTERNAL_DOWNLOADS]: The skill performs environment setup by installing Python dependencies via pip targeting a requirements file in an external sibling directory.
  • [PROMPT_INJECTION]: The skill ingests untrusted market data from external financial platforms (AKShare, EastMoney). It lacks explicit boundary markers or sanitization steps for this data, creating a surface for indirect prompt injection where malicious instructions could theoretically be embedded in financial data fields. Capability inventory: shell command execution for data fetching and file system access for caching.
  • [CREDENTIALS_UNSAFE]: Documentation references the use of an environment variable XUEQIU_TOKEN for specific tool authentication. While no secrets are hardcoded in the skill, it is designed to interact with sensitive API credentials.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 03:23 AM