> ## Documentation Index
> Fetch the complete documentation index at: https://docs.ntropii.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Microsoft Copilot Studio

> Wire the Ntro MCP server into a Microsoft Copilot Studio agent so it can read Ntropii state, trigger runs, and deploy workflows.

[Microsoft Copilot Studio](https://www.microsoft.com/en-us/microsoft-copilot/microsoft-copilot-studio) is the Power Platform low-code agent builder. It speaks MCP, which means a Copilot Studio agent can call the Ntro MCP server's tools, fetch its resources, and use its prompts — all from inside a Microsoft 365 / Power Platform tenant.

This is useful when you want fund-ops teams who already live in Teams and SharePoint to drive Ntropii without leaving their existing surface.

## Prerequisites

* A Microsoft 365 tenant with Copilot Studio licensed
* Maker access to a Power Platform environment

For most evaluations, point Copilot Studio at the [hosted test instance](/tooling/mcp-server#quick-start--hosted-test-instance) at `https://mcp.test.ntropii.com` — it's already an HTTPS endpoint, so no tunnel or container deploy is needed. Self-host only when you need a private Workspace or an environment you control end-to-end ([self-host options](/tooling/mcp-server#self-host-advanced)).

## Register the MCP server in Copilot Studio

<Steps>
  <Step title="Open the agent">
    In Copilot Studio, open (or create) the agent that should have Ntropii access.
  </Step>

  <Step title="Add an MCP connection">
    From the agent's settings, navigate to **Tools → Add tool → Model Context Protocol**.
  </Step>

  <Step title="Point it at the server">
    Server URL: `https://mcp.test.ntropii.com` (or your self-hosted URL).
    Authentication: leave the auth field as **none**. The hosted test instance currently runs without auth; self-hosted instances bake the API key in at server startup via `--connection` / `NTRO_API_KEY` rather than per-request headers.
  </Step>

  <Step title="Pick which tools to expose">
    Copilot Studio lets you whitelist specific MCP tools per agent. For a runbook authoring assistant, the typical set is `ntro_whoami`, `ntro_tenant_create`, `ntro_entity_create`, `ntro_runbook_list`, `ntro_workflow_create`, `ntro_workflow_run`, and (if external agents are involved) `ntro_agent_create`. Skip `ntro_integration_create` unless the agent should be able to register data platforms (which usually requires sensitive credentials you'd rather not put through a Copilot prompt).
  </Step>

  <Step title="Configure topics or actions">
    Wire the tools into your agent's topics. A common starter topic: "Run NAV for \${entity}" → calls `ntro_workflow_run` with the right payload, polls `ntro://tasks/{id}` for status, and reports back.
  </Step>

  <Step title="Test in the Test pane">
    Type "What tenants do I have access to?" in the test pane. The agent should call `ntro_whoami` (or fetch `ntro://tenants`) and reply with a list.
  </Step>
</Steps>

## Surfacing resources

Copilot Studio agents can use MCP **resources** as grounding context. Useful URIs:

* `ntro://tenants` — let the agent answer "what tenants exist?"
* `ntro://schedule` — let the agent answer "what NAV runs are pending?"
* `ntro://tasks/{id}` — let the agent answer "what's the status of run X?"

Configure these in the agent's **Knowledge** section so the planner can pull them when needed.

## Authentication and scoping (self-host only)

When you run your own MCP instance it runs with one API key. That key is tenant-scoped — so the Copilot Studio agent's surface area is limited to the tenant the key authorises. If you have three tenants, run three MCP server instances (each with its own key) and register them as three separate connections in Copilot Studio.

This is the right boundary: a marketing team's Copilot agent shouldn't be able to deploy workflows to the operations team's tenant just because they share an M365 tenant.

The hosted test instance has no auth and exposes only the test environment, so this scoping isn't a concern there — don't use it for anything you care about.

## Limitations to know about

* **No streaming output today.** Copilot Studio renders MCP tool results as a single block. Long-running operations (a NAV that takes 10 minutes) need to return quickly with a task ID, then the agent polls `ntro://tasks/{id}` on a follow-up turn.
* **No interactive HITL signals from inside Copilot Studio.** Workflows that pause for human review surface those reviews in the [Tenant UI](/get-started/tenant-architecture), not in Copilot Studio. The Copilot agent can tell the user "go to the Tenant UI to approve this step", but it can't render the approval form itself.
* **Copilot Studio's MCP support is evolving.** If a feature here looks different from what you see in your Copilot Studio UI, check the [Microsoft docs](https://learn.microsoft.com/en-us/microsoft-copilot-studio/) for the latest — the integration changed substantially in 2025.

## Related

<CardGroup cols={2}>
  <Card title="Claude Code" icon="square-terminal" href="/tooling/coding-agents/claude-code">
    Terminal-based coding agent — same MCP server.
  </Card>

  <Card title="MCP server reference" icon="plug" href="/tooling/mcp-server">
    All transports, tools, resources, and prompts.
  </Card>
</CardGroup>