langgraph-agent-patterns
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE] (SAFE): No malicious patterns detected. All analyzed files contain standard implementation patterns for the LangGraph library and do not exhibit any behaviors associated with exfiltration, persistence, or privilege escalation.
- [Category 8: Indirect Prompt Injection] (SAFE): The logic in several examples (router, supervisor) routes control flow based on message content. This represents an ingestion point for untrusted data. However, the skill provides no dangerous capabilities—such as file writes, network operations, or shell command execution—meaning there is no vulnerability surface for exploitation.
Audit Metadata