AI Agent Integrations
Runtime authorization for every major agent framework and LLM SDK. Two lines of code. No changes to your agent logic. Works with your existing stack.
How integration works
npm install veto-sdk or pip install vetoLLM SDK
Agent Framework
Tool-level authorization for LangChain agents and chains
State-aware authorization for LangGraph agent workflows
Multi-agent authorization and role-based tool access for CrewAI crews
Streaming-compatible guardrails for Vercel AI SDK agents
Type-safe authorization policies for PydanticAI agent tool calls
Multi-agent authorization for Microsoft AutoGen. Control tool access per agent and enforce policies on inter-agent tool delegation.
Runtime tool authorization for Amazon Bedrock agents. Complement IAM with per-call policies on Lambda action groups.
Browser Automation
Protocol
Core SDK
Related resources
The definitive guide to guardrail approaches and taxonomy
AI Agent Authorization GuideAuthentication vs authorization vs action scope control
Compare Guardrail ToolsFeature matrix of Veto, NeMo, Guardrails AI, Lakera, and more
Use Cases by IndustryFinance, healthcare, DevOps, enterprise, and more
Don't see your framework?
The core SDK works with any agent. We are adding native integrations regularly.