fine-tuning-expert

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill is a collection of educational resources and code snippets for machine learning engineers. It does not contain prompt injections, data exfiltration patterns, or obfuscated code.
  • [EXTERNAL_DOWNLOADS]: The provided code examples include functionality to download models and datasets from Hugging Face Hub (meta-llama, etc.) and utilities via Python's load_dataset and AutoModel.from_pretrained. These are standard, well-known services for the intended ML workflow.
  • [COMMAND_EXECUTION]: In references/deployment-optimization.md, the skill provides Python templates that use subprocess.run to call external tools like llama-quantize and convert_hf_to_gguf.py. These operations are correctly scoped to specific model quantization tasks and use environment variables for path management, representing standard developer tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 04:40 PM
Security Audit — agent-trust-hub — fine-tuning-expert