Integrations
Integrations
Add compliance to your existing AI stack.
AI Frameworks
LangChain
The most popular framework for building LLM applications.
Supported:
- Agents (ReAct, OpenAI Functions, Structured Chat)
- Chains (LCEL, legacy)
- Tools and retrievers
- All LLM providers
from protectron.langchain import ProtectronCallback
callback = ProtectronCallback(system_id="my-agent")
executor = AgentExecutor(agent=agent, tools=tools, callbacks=[callback])CrewAI
Multi-agent orchestration framework.
Supported:
- Multi-agent crews
- Task delegation
- Per-agent logging
- Inter-agent communication
from protectron.crewai import ProtectronCallback
callback = ProtectronCallback(system_id="my-crew")
crew = Crew(agents=[...], tasks=[...], callbacks=[callback])AutoGen
Microsoft's multi-agent conversation framework.
Supported:
- Agent conversations
- Code execution logging
- Human-in-the-loop
- Group chat
from protectron.autogen import ProtectronLogger
logger = ProtectronLogger(system_id="autogen-app")Vercel AI SDK
TypeScript SDK for AI applications.
Supported:
- streamText / generateText
- All model providers
- Edge runtime compatible
import { withProtectron } from '@protectron/vercel-ai';
const result = await withProtectron(
streamText({ model: openai('gpt-5.2'), prompt }),
{ systemId: 'my-app' }
);Custom Integration
Building something custom? Use our base SDK.
Python
from protectron import Protectron
protectron = Protectron(system_id="custom-app")
with protectron.trace("session-123") as trace:
trace.log_llm_call(model="gpt-5.2", input=[...], output="...")
trace.log_tool_call(tool="search", input={...}, output={...})TypeScript
import { Protectron } from '@protectron/sdk';
const protectron = new Protectron({ systemId: 'custom-app' });
await protectron.logEvent('llm_call', { model: 'gpt-5.2', ... });LLM Providers
Works with any LLM provider through our framework integrations:
| Provider | Status |
|---|---|
| OpenAI | ✅ Supported |
| Anthropic | ✅ Supported |
| Google (Gemini) | ✅ Supported |
| Azure OpenAI | ✅ Supported |
| AWS Bedrock | ✅ Supported |
| Mistral | ✅ Supported |
| Cohere | ✅ Supported |
| Ollama | ✅ Supported |
| Local models | ✅ Supported |
Coming Soon
| Integration | Status | ETA |
|---|---|---|
| Haystack | 🔄 In Development | Q1 2026 |
| LlamaIndex | 📋 Planned | Q1 2026 |
| Slack (notifications) | 📋 Planned | Q1 2026 |
| Jira (requirements sync) | 📋 Planned | Q2 2026 |
Want a specific integration? Let us know
Installation
Python
# Core SDK
pip install protectron
# With framework support
pip install protectron[langchain]
pip install protectron[crewai]
pip install protectron[autogen]
pip install protectron[all]TypeScript
npm install @protectron/sdk
npm install @protectron/vercel-aiGet Started
1
Create an account
Free 14-day trial
2
Install the SDK
Choose your framework
3
Add the callback
One line of code
4
View your traces
See events in your dashboard
Ready to Integrate?
Add EU AI Act compliance to your AI stack in minutes.
Need help with integration? Contact support@protectron.ai