grpo-rl-training
Warn
Audited by Snyk on Jun 23, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.75). The required runtime workflow loads an external dataset via
load_dataset("your/dataset")and passes its text into TRL training, so outsider-authored free-form dataset content can be ingested into the LLM context through the training prompts/completions.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata