skills/g1joshi/agent-skills/deepseek/Gen Agent Trust Hub

deepseek

Pass

Audited by Gen Agent Trust Hub on Oct 11, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the official DeepSeek API endpoint (https://api.deepseek.com) and its official GitHub organization. These are recognized as well-known service providers for AI model access and development.\n- [COMMAND_EXECUTION]: Includes instructions for self-hosting models using the vllm framework. The inclusion of the --trust-remote-code flag is a standard and documented requirement for executing the specific model architectures provided by the model vendor.\n- [CREDENTIALS_UNSAFE]: No hardcoded secrets were detected. The skill correctly demonstrates the use of environment variables (os.environ.get) and generic placeholders for API authentication.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 11, 2026, 03:36 AM