durable-state-patterns

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill contains exclusively educational content and code snippets for state management in AI agent frameworks like LangGraph. The provided examples use standard database connectors (SQLite, PostgreSQL, Redis) and established libraries (tenacity, langgraph, asyncpg). While database connection strings are shown in examples, they use generic placeholders (e.g., 'user:pass@localhost/db') rather than exposing sensitive credentials. The instructions focus on best practices for persistence, failure recovery, and state schema versioning.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 01:19 PM
Security Audit — agent-trust-hub — durable-state-patterns