langsmith-trace-analyzer

Warn

Audited by Snyk on Feb 14, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.80). The skill explicitly downloads LangSmith trace JSON (via scripts/download_traces.py using the langsmith-fetch CLI and the LangSmith SDK) and then parses/analyzes trace "messages" in scripts/analyze_traces.py, which are user-generated/untrusted runtime content (prompts, outputs, metadata) that the agent reads and interprets, enabling potential indirect prompt injection.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 14, 2026, 04:00 PM