skills/kumaran-is/claude-code-onboarding/llm-evaluation/Snyk

llm-evaluation

Warn

Audited by Snyk on Aug 6, 2026

Risk Level: MEDIUM

Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

Third-party content exposure detected (medium risk: 0.30). In the eval harness reference, the agent runtime consumes outsider-authored free text because it directly iterates over test_cases and sends case.input / user_message into the LangGraph/ADK agent (and can pass case.context into scoring), without selecting a specific pre-vetted item.

Issues (1)

W011

MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

Audit Metadata

Risk Level

MEDIUM

Analyzed

Aug 6, 2026, 01:54 PM

Issues

1

Security Audit — snyk — llm-evaluation