> SpecStream is the live web data engine for teams and agents. This is the Markdown copy of https://specstream.ai/docs/integrations/azure-api-center. Every page, with what SpecStream does and costs: https://specstream.ai/llms.txt

# Azure API Center

Register SpecStream's remote MCP server in your Azure API Center. Your developers find it in the API Center portal, in VS Code and in GitHub Copilot, and each one signs in with a SpecStream account of their own.

[SpecStream for Azure API CenterAPI & cloud platforms](https://specstream.ai/integrations/azure-api-center.md)

## What you need

-   An API center in your Azure subscription. Microsoft's quickstart, [Create an API center (opens in a new tab)](https://learn.microsoft.com/en-us/azure/api-center/set-up-api-center), sets one up.
-   Permission to add environments and assets in that API center.
-   A SpecStream account for each developer who uses the server. New accounts get $10 of credit on sign-up, no card. [Create one](https://specstream.ai/signup?utm_source=azure-api-center).

1.  1

    ## 1. Add an environment for SpecStream

    A remote MCP server in API Center belongs to an environment: the place where the server runs. SpecStream runs on its own host, so give it an environment of its own.

    In the [Azure portal (opens in a new tab)](https://portal.azure.com), open your API center. In the sidebar menu, under **Inventory**, select **Environments**, then **\+ New environment**. Set **Environment title** to `SpecStream`, keep the **Identification** API Center generates, and pick the **Environment type** your teams use for live services. Leave **Server Type** empty: SpecStream does not sit behind your API Management. Save the environment.

2.  2

    ## 2. Register SpecStream as an MCP server

    Under **Inventory**, select **Assets**, then **\+ Register an asset** > **MCP server**. In the **Register an MCP server** form, fill in the fields below.

    Under **Remotes**, select **\+ Add remote**, enter the **Runtime URL** `https://api.specstream.ai/mcp`, choose the `SpecStream` environment, and select **Save**. Then give the version a **Version title**, **Version identification** and **Version lifecycle**, add the **External documentation** link, and select **Create**.

    API Center then generates two definitions for the server version on its own: an SSE definition and a Streamable definition. SpecStream answers over Streamable HTTP.

    Register an MCP server

    ```text
    Title: SpecStream
    Summary: The live web data engine.
    Description: One platform to build, search, extract, enrich, map, and monitor data across the entire web.
    Icon URL: https://specstream.ai/favicon.svg
    Runtime URL: https://api.specstream.ai/mcp
    Environment: SpecStream
    Version title: SpecStream MCP
    Version identification: v1
    Version lifecycle: Production
    External documentation: https://specstream.ai/docs/mcp
    ```

3.  3

    ## 3. Publish it in the API Center portal

    If your API Center portal is not published yet, open **API Center portal** in the API center's menu, select **Configure Entra ID**, follow the **Quick setup** flow, then select **Save + publish**.

    In the portal, SpecStream appears among your MCP servers. Its page shows the endpoint, and a developer can install it in Visual Studio Code from there.

4.  4

    ## 4. Sign in and ask

    The first time VS Code starts the server, it asks the developer to sign in to SpecStream. They sign in, or create an account, and approve the access. No key is pasted and none is stored in the registry. Requests then run on that developer's account, with its balance and spending limits.

    In Copilot Chat agent mode, the developer asks in plain words and names SpecStream:

    In Copilot Chat

    ```text
    Use SpecStream to find the current price page of each vendor in vendors.csv, and add the plan names and monthly prices to the file.
    ```

## Make it the registry for GitHub Copilot

Enterprise and organization owners on Copilot Business or Copilot Enterprise can point Copilot at the API center, so developers see SpecStream next to your other approved servers.

First let Copilot read the registry: under **API Center portal**, in its settings, turn on anonymous access in the API visibility settings and select **Save + publish**. GitHub's own guide asks for this step.

Then, for an enterprise, open **AI controls** > **MCP** on GitHub, keep **MCP servers in Copilot** enabled, and enter the registry address in **MCP Registry URL**. For an organization, it is **Settings** > **Copilot** > **Policies**, field **MCP Registry URL (optional)**. Under **Restrict MCP access to registry servers**, **Registry only** lets only the servers in your API center run, and **Allow all** keeps every server open.

## Which registry address goes where

The two vendors write the address two ways, and the wrong one breaks the registry.

| Where | Address |
| --- | --- |
| GitHub **MCP Registry URL** | `https://<api-center-name>.data.<region>.azure-apicenter.ms/workspaces/default`, with no route after it. Copilot adds the `/v0.1/servers` path itself. |
| A client that reads the registry list directly | `https://<api-center-name>.data.<region>.azure-apicenter.ms/workspaces/default/v0.1/servers` |
| Never | The `portal` host, such as `https://<api-center-name>.portal.<region>.azure-apicenter.ms`. The registry lives on the `data` host. |

## What the server gives your developers

-   Search API through `search_web`: an answer from the live web with the pages it was read on, at $5 per 1,000 searches of up to 10 results.
-   Extract API through `read_pages`: the main text of up to 20 pages a call, at $0.80 per 1,000 pages. A page with no text is not charged.
-   Deep Search through `compile`, `quote_capacity_work` and `authorize_capacity_work`: a request in plain words returns every match as structured rows, at 5¢, 15¢ or $1 a row. The most it can cost is quoted first, and nothing runs until the developer approves it.
-   Source URL and read time on every value.

## Keys for shared agents

A service that cannot sign in, such as a build agent, uses an API key instead. In SpecStream, open [Agents & API](https://specstream.ai/workspace/agents?tab=keys&utm_source=azure-api-center), choose **API keys**, then **Create API key**, and give it a total spending limit. The client sends it as `Authorization: Bearer <key>` to `https://api.specstream.ai/mcp`. Keep the key in the client's own secret store, never in the API center.

## Cost

Registering SpecStream costs nothing on our side. GitHub's guide states that Azure API Center offers a free tier for basic cataloging and discovery, MCP registry management included. On SpecStream, reading and pricing a request are free, searches and pages come out of the balance as they run, and Deep Search never costs more than the maximum a developer approved. [Balance, quotes, and billing](https://specstream.ai/docs/billing.md) has the details.

## Remove it

Delete the SpecStream asset under **Inventory** > **Assets** to take it out of the portal and the registry. A developer ends VS Code's access on the SpecStream side under **Agents & API** > **OAuth applications**, and a shared key is revoked under **API keys**. Saved results stay in the workspace. Questions go to [support@specstream.ai](mailto:support@specstream.ai).

## More

[SpecStream for Azure API Center](https://specstream.ai/integrations/azure-api-center.md) covers what it does in Azure API Center on one page. Every other tool has its guide in [Connect your tools](https://specstream.ai/docs/integrations.md). You see the most a request can cost before it starts, and the matches are paid from your balance. [Billing](https://specstream.ai/docs/billing.md) has the details. For help, email [support@specstream.ai](mailto:support@specstream.ai).
