Framework Integration Overview

AxonFlow integrates with popular AI and ML frameworks to add governance, policy enforcement, and audit logging to your existing AI applications. Add compliance guardrails without rewriting your agent code.

Supported Frameworks

LLM Orchestration Frameworks

Framework	Language	Integration Type	Documentation
LangChain	Python	SDK + Raw HTTP	Most comprehensive guide
LangChainGo	Go	Go SDK	Native Go integration
LlamaIndex.TS	TypeScript	TypeScript SDK	Node.js/TypeScript apps
CrewAI	Python	Python SDK	Multi-agent crews
Lyzr	Python	Python SDK	Enterprise AI agents

Enterprise Platforms

Platform	Integration Type	Use Case	Documentation
Microsoft Copilot Studio	HTTP API	Low-code AI apps	Power Platform integration
Obot	Go SDK	Kubernetes-native agents	Cloud-native deployment

Integration Patterns

Gateway Mode

The recommended pattern for existing framework integrations (LangChain, CrewAI, LlamaIndex):

Your Framework → AxonFlow Pre-check → LLM Provider → AxonFlow Audit

Benefits:

• Policy enforcement before LLM calls
• Complete audit trail of all operations
• Token usage and cost tracking
• Works with any LLM provider

Proxy Mode

For simpler integrations:

Your Framework → AxonFlow Proxy → LLM Provider

Benefits:

• Single API endpoint
• Automatic policy enforcement
• Simpler integration (one call)

SDK Coverage

SDK	Frameworks Using It
Python SDK	LangChain, CrewAI, Lyzr
Go SDK	LangChainGo, Obot
TypeScript SDK	LlamaIndex.TS, Lyzr
Raw HTTP	All frameworks (reference implementation)

Quick Start by Framework

Python Frameworks (LangChain, CrewAI, Lyzr)

from axonflow import AxonFlow

async with AxonFlow(
    agent_url="http://localhost:8080",
    api_key="your-api-key"
) as client:
    # Pre-check before LLM call
    ctx = await client.get_policy_approved_context(
        user_id="user-123",
        query="Your prompt here"
    )

    if ctx.approved:
        # Make your framework's LLM call here
        response = await your_framework_llm_call(ctx.query)

        # Audit the result
        await client.audit_llm_call(
            context_id=ctx.context_id,
            provider="openai",
            model="gpt-4",
            response=response
        )

Go Frameworks (LangChainGo, Obot)

import "github.com/getaxonflow/axonflow-sdk-go"

client := axonflow.NewClient("http://localhost:8080", "your-api-key")

// Pre-check
ctx, _ := client.GetPolicyApprovedContext(context.Background(), &axonflow.PolicyRequest{
    UserID: "user-123",
    Query:  "Your prompt here",
})

if ctx.Approved {
    // Your framework LLM call
    response := yourFrameworkLLMCall(ctx.Query)

    // Audit
    client.AuditLLMCall(context.Background(), &axonflow.AuditRequest{
        ContextID: ctx.ContextID,
        Provider:  "openai",
        Model:     "gpt-4",
        Response:  response,
    })
}

TypeScript Frameworks (LlamaIndex.TS)

import { AxonFlow } from '@axonflow/sdk';

const client = new AxonFlow({
  agentUrl: 'http://localhost:8080',
  apiKey: 'your-api-key'
});

// Pre-check
const ctx = await client.getPolicyApprovedContext({
  userId: 'user-123',
  query: 'Your prompt here'
});

if (ctx.approved) {
  // Your framework LLM call
  const response = await yourFrameworkLLMCall(ctx.query);

  // Audit
  await client.auditLLMCall({
    contextId: ctx.contextId,
    provider: 'openai',
    model: 'gpt-4',
    response
  });
}

What AxonFlow Adds

Capability	Description
PII Detection	12+ PII types automatically detected and optionally redacted
SQL Injection Scanning	37+ attack patterns blocked in prompts and responses
Policy Enforcement	Custom rules in Rego/OPA with single-digit ms evaluation
Audit Logging	Complete request/response logging with compliance retention
Cost Tracking	Token usage and cost per request
Multi-Model Routing	Route to different LLM providers based on policy

Community vs Enterprise

Feature	Community	Enterprise
Framework integrations	All	All
PII detection	✅	✅
SQL injection (basic)	✅	✅
SQL injection (advanced)		✅
Audit logging	✅	✅
Policy enforcement	✅	✅
HITL approval queue		✅
Compliance exports		✅
Customer Portal		✅

Choosing a Framework Guide

If you're using...	Start with...
Python + need multi-agent	CrewAI Guide
Python + RAG focus	LangChain Guide
Go backend	LangChainGo Guide
TypeScript/Node.js	LlamaIndex.TS Guide
Low-code/Power Platform	Copilot Studio Guide
Kubernetes-native	Obot Guide

Need a Different Framework?

AxonFlow's SDK and HTTP APIs work with any framework. Use the LangChain guide as a reference - it includes both SDK and raw HTTP examples that can be adapted to any framework.

For framework-specific integration help:

• Community: GitHub Discussions
• Enterprise: [email protected]