Agent Framework

Synero + LangChain

Add Synero as a custom tool in your LangChain agent. Your agent can query four AI models simultaneously and get a structured JSON response with individual perspectives and a unified synthesis.

Setup

Get started in minutes

1

Install dependencies

Install the required Python packages.

pip install requests langchain
2

Create the Synero tool

Create a LangChain tool that calls the Synero API in JSON mode.

import requests
from langchain.tools import tool

@tool
def query_synero_council(prompt: str) -> dict:
    """Query 4 AI models simultaneously and get a synthesized answer.
    Use this when you need high-confidence answers or want multiple
    perspectives on a question."""
    response = requests.post(
        "https://synero.ai/api/query",
        headers={
            "Authorization": "Bearer sk_live_your_key_here",
            "Content-Type": "application/json",
        },
        json={"prompt": prompt, "format": "json"},
    )
    return response.json()
3

Add to your agent

Add the Synero tool to your LangChain agent's toolkit.

from langchain.agents import create_tool_calling_agent

tools = [query_synero_council]
agent = create_tool_calling_agent(llm, tools, prompt)

Full Example

Complete integration

import requests
from langchain.tools import tool

@tool
def query_synero_council(prompt: str) -> dict:
    """Query 4 AI models simultaneously and get a synthesized answer.
    Use this when you need high-confidence answers or want multiple
    perspectives on a question."""
    response = requests.post(
        "https://synero.ai/api/query",
        headers={
            "Authorization": f"Bearer {os.environ['SYNERO_API_KEY']}",
            "Content-Type": "application/json",
        },
        json={"prompt": prompt, "format": "json"},
    )
    return response.json()

FAQ

Does Synero work with LangGraph?

Yes. The Synero tool works with any LangChain-compatible agent framework including LangGraph. Just add the tool to your agent's toolkit.

What does the JSON response look like?

The response includes individual advisor responses (with model name and latency), the synthesized answer, and token usage. Each advisor has a unique persona: Architect, Philosopher, Explorer, and Maverick.

Related

Start using Synero with LangChain

Four AI models. One synthesized answer. Integrated into your workflow.