evaluation-harness

Warn

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: MEDIUMPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION] (MEDIUM): The score_with_llm function in SKILL.md implements an 'LLM as judge' pattern that is vulnerable to indirect prompt injection.
  • Ingestion points: It processes untrusted model outputs (actual) and dataset entries (expected) which are interpolated into the prompt.
  • Boundary markers: No delimiters or isolation techniques are used in the prompt template to separate data from instructions.
  • Capability inventory: The resulting scores can trigger sys.exit(1), potentially failing automated builds and affecting the deployment process.
  • Sanitization: The input is interpolated directly into the judge's prompt without any sanitization or validation of its content.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 08:48 PM