A Model Context Protocol server that brings Azure DevOps context to AI assistants. Browse projects, work items, repositories, pipelines, wikis, and more using natural language — powered by your own Microsoft identity.
organization parameter. Best for general-purpose clients.{ "status": "ok", "version": "..." }.Add a .vscode/mcp.json file to your project:
{
"servers": {
"ado": {
"type": "http",
"url": "https://ado-mcp.liventus.com/org/{your-org}/mcp"
}
}
}
Replace {your-org} with your Azure DevOps organization name (e.g. Contoso).
{
"servers": {
"ado": {
"type": "http",
"url": "https://ado-mcp.liventus.com/mcp"
}
}
}
Add a .mcp.json file to your project:
{
"mcpServers": {
"ado": {
"type": "http",
"url": "https://ado-mcp.liventus.com/org/{your-org}/mcp"
}
}
}
Replace {your-org} with your Azure DevOps organization name (e.g. Contoso).
{
"mcpServers": {
"ado": {
"type": "http",
"url": "https://ado-mcp.liventus.com/mcp"
}
}
}
With the multi-org endpoint you can say "List my open tasks for Contoso" — the LLM calls
ado_list_organizations, matches the name, then queries work items with the correct org identifier.
Authentication is handled via Microsoft Entra ID OAuth. No personal access tokens or environment variables are required for the hosted server — just sign in with your Microsoft account and your existing Azure DevOps permissions apply automatically.
Full documentation is available at https://ado-mcp.liventus.com/README.md.