Klavis AI

What is Klavis AI about?

Klavis AI delivers a catalog of Model Context Protocol (MCP) servers that expose the functionality of popular SaaS platforms (GitHub, Gmail, Google Sheets, Slack, Notion, Salesforce, Postgres, etc.) as AI‑friendly endpoints. Both a managed hosted service and self‑hostable Docker images are available, allowing developers to plug external data and actions directly into LLM‑driven agents.

How to use Klavis AI?

1. Hosted service (quickest) – Sign up for a free API key, install the klavis client (pip install klavis or npm install klavis), and call klavis.mcp_server.create_server_instance("GMAIL", "user123") to obtain a ready‑to‑use server URL.
2. Self‑hosting – Pull a Docker image, e.g., docker run -p 5000:5000 ghcr.io/klavis-ai/github-mcp-server:latest, or build from source under klavis/mcp_servers/<service>.
3. Integrate with LLMs – Retrieve the server's tool definitions in OpenAI function‑calling format and pass them to the model as part of a chat completion request.

Key features of Klavis AI

• 50+ ready‑made integrations covering email, version control, spreadsheets, video platforms, CRM, databases, and more.
• Enterprise‑grade hosted infrastructure with 99.9 % SLA, SOC2/GDPR compliance, and scalable auto‑scaling.
• OAuth‑handled servers that abstract away token management for Google, GitHub, Slack, Salesforce, etc.
• Multi‑language client libraries (Python, TypeScript) for rapid prototyping.
• Docker images for every server and source code for full customization.
• Zero‑config Cursor/Claude Desktop integration via generated JSON snippets.

Use cases of Klavis AI

• AI‑powered assistants that read/write Gmail, create GitHub issues, or update Google Sheets on demand.
• Automated workflow bots that sync data between Slack and Notion, or trigger Salesforce actions from LLM prompts.
• Prototype and production pipelines where developers start with the hosted service and later migrate to self‑hosted servers for compliance.
• Function‑calling demonstrations for teaching LLM tool usage with real‑world SaaS APIs.

FAQ from the Klavis AI

Q: Do I need to manage OAuth credentials myself? A: Not when using the hosted service – the platform handles the OAuth flow. For self‑hosted servers, set the required env vars (e.g., KLAVIS_API_KEY) and follow each server’s README.

Q: Is there a free tier? A: Yes, a free API key provides limited monthly usage for the hosted MCP service; additional usage is pay‑as‑you‑go.

Q: Can I add my own MCP server? A: Absolutely. Fork the repository, implement the MCP spec for your service, and publish a Docker image or contribute back.

Q: What languages are supported? A: The client SDK is available for Python and TypeScript/JavaScript. Server implementations exist in Go, Python, and Node.js.

Q: How do I retrieve tool definitions for OpenAI? A: Call klavis.mcp_server.list_tools(server_url, format="OPENAI") to get a JSON schema compatible with OpenAI’s function‑calling feature.

🎯 Choose Your Solution

Quick Start

Option 1: Cloud-hosted - klavis.ai

Quickstart guide →

Option 2: Self-host

# Run any MCP Integration
docker pull ghcr.io/klavis-ai/github-mcp-server:latest
docker run -p 5000:5000 ghcr.io/klavis-ai/github-mcp-server:latest

# Install Open Source Strata locally
pipx install strata-mcp
strata add --type stdio playwright npx @playwright/mcp@latest

Option 3: SDK

# Python SDK
from klavis import Klavis
from klavis.types import McpServerName

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

# Create Strata instance
strata = klavis_client.mcp_server.create_strata_server(
    user_id="user123",
    servers=[McpServerName.GMAIL, McpServerName.SLACK],
)

# Or use individual MCP servers
gmail = klavis.mcp_server.create_server_instance(
    server_name=McpServerName.GMAIL,
    user_id="user123",
)

// TypeScript SDK
import { KlavisClient, McpServerName } from 'klavis';

const klavis = new KlavisClient({ apiKey: 'your-api-key' });

// Create Strata instance
const strata = await klavis.mcpServer.createStrataServer({
    userId: "user123",
    servers: [Klavis.McpServerName.Gmail, Klavis.McpServerName.Slack],
});

// Or use individual MCP servers
const gmail = await klavis.mcpServer.createServerInstance({
    serverName: McpServerName.GMAIL,
    userId: "user123"
});

Option 4: REST API

# Create Strata server
curl -X POST "https://api.klavis.ai/v1/mcp-server/strata" \
  -H "Authorization: Bearer your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "user_id": "user123",
    "servers": ["GMAIL", "SLACK"]
  }'

# Create individual MCP server
curl -X POST "https://api.klavis.ai/v1/mcp-server/instance" \
  -H "Authorization: Bearer your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "server_name": "GMAIL",
    "user_id": "user123"
  }'

Resources

• 📖 Documentation
• 💬 Discord Community
• 🐛 Report Issues
• 🌐 Klavis AI Website