deepseek-balance

Pass

Audited by Gen Agent Trust Hub on Aug 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes python scripts/check_balance.py to perform the balance check. Additionally, on Windows systems, the script invokes chcp 65001 via subprocess.run to ensure the terminal uses UTF-8 encoding.
  • [EXTERNAL_DOWNLOADS]: The Python script makes a network request to the well-known service endpoint https://api.deepseek.com/user/balance. This is the legitimate destination for the skill's intended purpose.
  • [PROMPT_INJECTION]: The skill processes data from an external API, presenting an indirect prompt injection surface.
  • Ingestion points: scripts/check_balance.py reads JSON response data from the DeepSeek API.
  • Boundary markers: No explicit delimiters are used in the script's stdout to separate the API response from other context.
  • Capability inventory: The skill possesses network access and file system read capabilities for configuration files (.env).
  • Sanitization: The script outputs the raw body of the API response to the agent without additional filtering or escaping.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 8, 2026, 02:07 AM
Security Audit — agent-trust-hub — deepseek-balance