binance-account

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external Binance API endpoints, creating a surface where malicious instructions could be injected via account metadata or trade history.
  • Ingestion points: API responses fetched in scripts/account_scenarios.py and scripts/bn_account.py via python-binance client methods (e.g., get_account, futures_account, get_my_trades).
  • Boundary markers: The skill does not implement delimiters or specific instructions to ignore potential commands embedded within the API data.
  • Capability inventory: The skill is restricted to read-only Binance API operations and console output; it does not possess file system write access or arbitrary command execution capabilities.
  • Sanitization: No validation or escaping is applied to the data retrieved from Binance before it is output to the agent's context.
  • [CREDENTIALS_UNSAFE]: The skill includes hardcoded authentication details for an internal proxy service.
  • Evidence: The variable HK_PROXY is defined as "http://hk:x@sc-vpn.internal:8080" in SKILL.md, scripts/account_scenarios.py, and scripts/bn_account.py, exposing a username and password for the internal VPN proxy.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 04:28 AM
Security Audit — agent-trust-hub — binance-account