Connect Your AI Client
Connect the IMVASA MCP server to Claude Desktop, Cursor, ChatGPT, Windsurf, Zed, VS Code extensions, or custom automated agents. Find your environment below for copy-ready configuration files and setup steps.
1. Claude Desktop
HTTP TransportOpen Claude Desktop Settings → Developer → Edit Config, or edit your configuration file directly:
~/Library/Application Support/Claude/claude_desktop_config.json
%APPDATA%\Claude\claude_desktop_config.json
~/.config/Claude/claude_desktop_config.json
{
"mcpServers": {
"imvasa-audit": {
"url": "https://seoaudit.imvasa.dev/mcp"
}
}
}Restart Claude Desktop after saving. You'll see the IMVASA Audit Tool tools appear in the tool picker.
2. Cursor IDE
mcp.jsonAdd to ~/.cursor/mcp.json (globally) or .cursor/mcp.json in your project:
{
"mcpServers": {
"imvasa-audit": {
"url": "https://seoaudit.imvasa.dev/mcp"
}
}
}3. ChatGPT (Custom Connector)
Web / Desktop- Open ChatGPT Settings → Connectors / Developer Tools → Add Connector.
- Paste the server URL:
https://seoaudit.imvasa.dev/mcp - Save without authentication headers.
- Enable the connector in a new chat and ask it to audit a URL.
4. Windsurf (Codeium)
mcp_config.jsonAdd to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"imvasa-audit": {
"serverUrl": "https://seoaudit.imvasa.dev/mcp"
}
}
}5. Zed Editor
settings.jsonIn Zed user settings.json, add under context_servers:
{
"context_servers": {
"imvasa-audit": {
"url": "https://seoaudit.imvasa.dev/mcp"
}
}
}6. VS Code (Cline / Roo Code / Continue)
MCP ExtensionConfigure your MCP extension with the following definition:
{
"mcpServers": {
"imvasa-audit": {
"url": "https://seoaudit.imvasa.dev/mcp",
"disabled": false
}
}
}7. Programmatic SDKs (Node.js & Python)
SDK CodeConnect custom automated agents, CI/CD scripts, or background workers using the official TypeScript and Python SDKs:
import { Client } from "@modelcontextprotocol/sdk/client/index.js";
import { StreamableHttpClientTransport } from "@modelcontextprotocol/sdk/client/streamableHttp.js";
const transport = new StreamableHttpClientTransport(new URL("https://seoaudit.imvasa.dev/mcp"));
const client = new Client({ name: "my-agent", version: "1.0.0" }, { capabilities: {} });
await client.connect(transport);
const result = await client.callTool({
name: "run_audit",
arguments: { url: "https://example.com" }
});import asyncio
from mcp import ClientSession
from mcp.client.streamable_http import streamablehttp_client
async def run():
async with streamablehttp_client("https://seoaudit.imvasa.dev/mcp") as (read, write):
async with ClientSession(read, write) as session:
await session.initialize()
res = await session.call_tool("find_issues", arguments={"url": "https://example.com"})
print(res.content[0].text)
asyncio.run(run())8. Generic HTTP Clients & Other Tools
HTTP StreamAny MCP client that supports Streamable HTTP transport can connect. Point it at the server URL above. No authentication headers are required.
Accept: application/json, text/event-stream
Content-Type: application/json