Skip to main content

​
Installation

pip install klavis

​
Get Your API Key

Sign up at klavis.ai and create your API key.

​
Quick Start

from klavis import Klavis
from klavis.types import McpServerName

klavis_client = Klavis(api_key="your-klavis-key")

# Create a YouTube MCP server instance
youtube_server = klavis_client.mcp_server.create_server_instance(
    server_name=McpServerName.YOUTUBE,
    user_id="user123", # Change to user id in your platform
)

print(f"Server created: {youtube_server.server_url}")

​
Integration with MCP Client

If you already have an MCP client implementation in your codebase: 
from klavis import Klavis
from klavis.types import McpServerName

klavis_client = Klavis(api_key="your-klavis-key")

# Create a YouTube MCP server instance
youtube_server = klavis_client.mcp_server.create_server_instance(
    server_name=McpServerName.YOUTUBE,
    user_id="user123"
)

print(f"Server created: {youtube_server.server_url}")

​
Function Calling with OpenAI

Integrate directly with OpenAI using function calling: 
import json
from openai import OpenAI
from klavis import Klavis
from klavis.types import McpServerName, ToolFormat

OPENAI_MODEL = "gpt-4o-mini"

openai_client = OpenAI(api_key="YOUR_OPENAI_API_KEY")
klavis_client = Klavis(api_key="YOUR_KLAVIS_API_KEY")

# Create server instance
youtube_server = klavis_client.mcp_server.create_server_instance(
    server_name=McpServerName.YOUTUBE,
    user_id="user123"
)

# Get available tools in OpenAI format
tools = klavis_client.mcp_server.list_tools(
    server_url=youtube_server.server_url,
    format=ToolFormat.OPENAI
)

# Initial conversation
messages = [{"role": "user", "content": "Summarize this video: https://youtube.com/watch?v=..."}]

# First OpenAI call with function calling
response = openai_client.chat.completions.create(
    model=OPENAI_MODEL,
    messages=messages,
    tools=tools.tools
)

messages.append(response.choices[0].message)

# Handle tool calls
if response.choices[0].message.tool_calls:
    for tool_call in response.choices[0].message.tool_calls:
        result = klavis_client.mcp_server.call_tools(
            server_url=youtube_server.server_url,
            tool_name=tool_call.function.name,
            tool_args=json.loads(tool_call.function.arguments)
        )
        
        # Add tool result to conversation
        messages.append({
            "role": "tool",
            "tool_call_id": tool_call.id,
            "content": str(result)
        })

# Second OpenAI call to process tool results and generate final response
final_response = openai_client.chat.completions.create(
    model=OPENAI_MODEL,
    messages=messages
)

print(final_response.choices[0].message.content)

​
Authentication

​
OAuth Services

For OAuth services like Gmail, Google Drive, etc.: 
# Create server instance for OAuth service
server = klavis_client.mcp_server.create_server_instance(
    server_name=McpServerName.GMAIL,
    user_id="user123"
)

# OAuth URL is provided in server.oauth_url
import webbrowser
webbrowser.open(server.oauth_url)

​
API Key Services

For services that require API keys: 
# Set authentication token for API key services
klavis_client.mcp_server.set_auth_token(
    instance_id=server.instance_id,
    auth_token="your-service-api-key"
)

​
Multi-Tool Workflows

Combine multiple MCP servers for complex workflows: 
# Create multiple servers
github_server = klavis_client.mcp_server.create_server_instance(
    server_name=McpServerName.GITHUB,
    user_id="user123"
)

slack_server = klavis_client.mcp_server.create_server_instance(
    server_name=McpServerName.SLACK,
    user_id="user123"
)

# Use tools from both servers in a single AI conversation
all_tools = []
all_tools.extend(klavis_client.mcp_server.list_tools(github_server.server_url).tools)
all_tools.extend(klavis_client.mcp_server.list_tools(slack_server.server_url).tools)

# Initialize conversation
messages = [{"role": "user", "content": "Create a GitHub issue and notify the team on Slack"}]

# Loop to let LLM work with multiple tools
max_iterations = 5
for iteration in range(max_iterations):
    response = openai_client.chat.completions.create(
        model="gpt-4",
        messages=messages,
        tools=all_tools
    )
    
    messages.append(response.choices[0].message)
    
    # Check if LLM wants to use tools
    if response.choices[0].message.tool_calls:
        for tool_call in response.choices[0].message.tool_calls:
            # Determine which server to use based on tool name
            server_url = github_server.server_url if "github" in tool_call.function.name else slack_server.server_url
            
            # Execute tool
            result = klavis_client.mcp_server.call_tools(
                server_url=server_url,
                tool_name=tool_call.function.name,
                tool_args=json.loads(tool_call.function.arguments)
            )
            
            # Add tool result to conversation
            messages.append({
                "role": "tool",
                "tool_call_id": tool_call.id,
                "content": str(result)
            })
    else:
        # LLM finished the task
        print(f"Task completed in {iteration + 1} iterations")
        print(response.choices[0].message.content)
        break

​
Error Handling

from klavis.exceptions import KlavisError

try:
    server = klavis_client.mcp_server.create_server_instance(
        server_name=McpServerName.YOUTUBE,
        user_id="user123"
    )
except KlavisError as e:
    print(f"Error creating server: {e}")

​
Next Steps

API Reference

Complete API documentation

MCP Servers

Explore available MCP servers
⌘I