About
Lokka is an MCP server that lets you query and manage your Azure and Microsoft 365 tenants using natural language through the Microsoft Graph and Azure Resource Manager APIs. It supports interactive, app‑only, or client‑provided token authentication for flexible deployment.
Capabilities

Lokka is a Model Context Protocol (MCP) server that bridges AI assistants with the Microsoft Graph and Azure Resource Manager APIs. By exposing a rich set of tools, prompts, and sampling capabilities, it lets developers turn natural‑language requests into authenticated calls against Azure and Microsoft 365 tenants. In practice, this means a Claude Desktop user can ask for complex tenant queries—such as listing conditional‑access policies that exclude emergency accounts or creating a dynamic security group—and Lokka will translate those requests into the appropriate Microsoft API calls, handling authentication and response formatting automatically.
The server solves a core pain point for cloud‑native teams: integrating the breadth of Microsoft’s management APIs into conversational workflows without writing boilerplate code. Developers no longer need to write custom OAuth flows, manage SDKs, or parse raw Graph responses; Lokka provides a single point of contact that understands both the MCP protocol and Microsoft’s authentication models. It supports interactive user login, app‑only certificate or client‑secret flows, and a client‑provided token mode—giving flexibility for on‑premises, CI/CD, or user‑direct scenarios.
Key features include:
- Unified authentication: Multiple secure options (interactive, certificate‑based, client secret, token injection) ensure that Lokka can run in any deployment context.
- Rich toolset: Built‑in tools for token management (, ) and a growing list of Microsoft‑specific operations (security groups, conditional access, Intune policies, cost analysis).
- Natural language understanding: The MCP client can send plain‑English prompts; Lokka maps them to API calls, returning structured data that the assistant can embed in its responses.
- Extensibility: Developers can add custom prompts or sampling strategies, tailoring the assistant’s behavior to their organization’s policies.
Typical use cases span IT operations, compliance audits, and cost optimization. An IT administrator can ask the assistant to “show me all Intune device policies for the Call Center group,” and Lokka will retrieve and format that data. A finance team might query “what was the most expensive Azure service last month?” and receive a concise cost breakdown. In automated pipelines, Lokka can create or modify security groups based on business rules, enabling zero‑touch governance.
By integrating with MCP clients like Claude Desktop or any MCP‑compatible chat interface, Lokka turns conversational AI into a powerful tenant management tool. Its ability to translate human intent into secure, authenticated API calls—while abstracting the complexities of Microsoft’s authentication mechanisms—makes it a standout solution for developers looking to embed cloud management directly into AI workflows.
Related Servers
MarkItDown MCP Server
Convert documents to Markdown for LLMs quickly and accurately
Context7 MCP
Real‑time, version‑specific code docs for LLMs
Playwright MCP
Browser automation via structured accessibility trees
BlenderMCP
Claude AI meets Blender for instant 3D creation
Pydantic AI
Build GenAI agents with Pydantic validation and observability
Chrome DevTools MCP
AI-powered Chrome automation and debugging
Weekly Views
Server Health
Information
Explore More Servers
Deepwiki MCP Server
Unofficial Deepwiki crawler to Markdown
Python Pip MCP Server
Minimal MCP client and server for Anthropic models in Python
Pinner MCP Server
Pin third‑party dependencies to immutable digests with ease.
Windows MCP
AI-driven Windows UI automation without vision
StockSense MCP Server
Modular finance insights via lightweight microservices
NugetMcpServer
Real‑time NuGet API definitions for AI and dev tools