---
title: "Azure OpenAI"
description: "Create or reuse an Azure OpenAI resource and model deployment in Azure AI Foundry. From Keys and Endpoint copy the resource’s HTTPS origin and an API key into this connection; model deployment names are supplied per comm"
---

> Documentation Index
> Fetch the complete documentation index at: https://help.tooluse.net/llms.txt
> Use this file to discover all available pages before exploring further.

# Azure OpenAI

<p>Create or reuse an Azure OpenAI resource and model deployment in Azure AI Foundry. From Keys and Endpoint copy the resource’s HTTPS origin and an API key into this connection; model deployment names are supplied per command. This Azure-resource integration is separate from OpenAI platform keys. It pins the supported 2024-10-21 API and non-streaming JSON, with explicit output/input bounds. Model calls consume your Azure quota and charges; set resource spending controls and never automatically retry an uncertain request. One explicit request/page per invocation; no retries, automatic pagination or background work. This is a documented common-command subset, not the full provider API.</p>

<details data-command-index><summary>What you can do (3)</summary>
<h2>What you can do</h2>
<p>Available commands: 3 (1 read, 2 write). Enable only the commands your application needs.</p>
<div class="nb-table-scroll"><table>
<thead>
<tr>
<th>Command</th>
<th>Effect</th>
</tr>
</thead>
<tbody><tr>
<td><a href="https://help.tooluse.net/integrations/azure-openai#command-azure-openai-models-list">List models available to this Azure resource</a> — <code>azure-openai.models.list</code></td>
<td>Read</td>
</tr>
<tr>
<td><a href="https://help.tooluse.net/integrations/azure-openai#command-azure-openai-chat-create">Create one bounded non-streaming chat response</a> — <code>azure-openai.chat.create</code></td>
<td>Write</td>
</tr>
<tr>
<td><a href="https://help.tooluse.net/integrations/azure-openai#command-azure-openai-embeddings-create">Embed one explicit text input</a> — <code>azure-openai.embeddings.create</code></td>
<td>Write</td>
</tr>
</tbody></table>
</div>
</details>
<h2>Connect and use</h2>

<p><a href="https://www.tooluse.net/tools/integrations/azure-openai">Connect and use</a></p>
<p>Open the integration&#39;s connection form to check the available credential source. The source has not been resolved for this documentation export. The instructions below apply when using your own provider account.</p>
<h2>Get your provider credentials</h2>
<p>Follow these steps for Azure OpenAI, then enter the issued values in the connection form below.</p>
<ol>
<li>Create or reuse an Azure OpenAI resource and model deployment in Azure AI Foundry. From Keys and Endpoint copy the resource’s HTTPS origin and an API key into this connection; model deployment names are supplied per command. This Azure-resource integration is separate from OpenAI platform keys. It pins the supported 2024-10-21 API and non-streaming JSON, with explicit output/input bounds. Model calls consume your Azure quota and charges; set resource spending controls and never automatically retry an uncertain request.</li>
</ol>
<p><a href="https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/create-resource">Official credential setup</a></p>
<h2>Where to enter credentials</h2>
<p>In the integration page, sign in, choose your workspace and open Connect. If a <strong>Credentials</strong> choice is shown, select <strong>Use my credentials</strong>. Fill the following connection fields and save. Provider credentials belong in this form, never in an MCP configuration or tool input.</p>
<div class="nb-table-scroll"><table>
<thead>
<tr>
<th>Connection field</th>
<th>Requirement</th>
</tr>
</thead>
<tbody><tr>
<td>Azure OpenAI HTTPS site origin</td>
<td>Required</td>
</tr>
<tr>
<td>Provider API token</td>
<td>Required</td>
</tr>
</tbody></table>
</div><p>Open API keys and create a Tooluse.net key for the enabled commands.</p>
<p>Set TOOLUSE_API_KEY in your client and use the shared MCP configuration.</p>
<p><a href="https://help.tooluse.net/clients">Configure Claude Code, Codex or Cursor once</a></p>

<h2>Permissions and results</h2>

<p>If you have more than one workspace, check the selected workspace in the connection or API-key dialog before saving. Roles, enabled commands, expiry and revocation use Tooluse.net&#39;s existing permissions. Disabling a connection or revoking a key immediately prevents new calls.</p>

<p>Send one Claude tool_use block per request. Keep the same Idempotency-Key when retrying the same request; use a new key for a new call. The response includes the shared Tooluse.net receipt and Claude tool_result. Check both HTTP status and the returned tool result before treating a call as successful.</p>

<p>Your connection&#39;s daily allowance and key permissions also apply. Adding a connection does not make a provider request.</p>

<p>Calls return the shared execution receipt and a Claude tool_result. Provider output varies by operation; consult the provider documentation for its response fields. Unconfirmed calls should be checked before retrying.</p>

<h2>Call from your client</h2>

<p>Use the shared MCP configuration. Refresh your client&#39;s tool list after saving the connection or changing key permissions, then call the returned tool name with the example input below.</p>

<p><a href="https://help.tooluse.net/clients">Configure Claude Code, Codex or Cursor once</a></p>

<h2>Discover your enabled tools</h2>

<p>List permitted tools in Claude format. Read every page using page and size (maximum 100) until all results have been received. Use the returned name, not a catalog capability identifier, in the tool_use block.</p>

<pre><code class="language-bash">curl 'https://api.tooluse.net/v1/tools' \
  -H "Authorization: Bearer $TOOLUSE_API_KEY"
</code></pre>

<p>Find this command in the tool list permitted by your key. Replace svc_FROM_YOUR_TOOL_LIST with its returned name. Use a new Idempotency-Key for each new call and retain it when retrying that same call.</p>

<h2>HTTP call</h2>

<pre><code class="language-bash">curl 'https://api.tooluse.net/v1/tool-calls' \
  -H "Authorization: Bearer $TOOLUSE_API_KEY" \
  -H 'Content-Type: application/json' \
  -H 'Idempotency-Key: replace-with-a-new-call-id' \
  --data '{
  "type": "tool_use",
  "id": "toolu_example",
  "name": "svc_FROM_YOUR_TOOL_LIST",
  "input": {
    "query": {
      "api-version": "2024-10-21"
    }
  }
}'
</code></pre>

<p>Send one Claude tool_use block per request. Keep the same Idempotency-Key when retrying the same request; use a new key for a new call. The response includes the shared Tooluse.net receipt and Claude tool_result. Check both HTTP status and the returned tool result before treating a call as successful.</p>

<h2>Commands</h2>

<details id="command-azure-openai-models-list">
<summary><span>List models available to this Azure resource — <code>azure-<wbr>openai.<wbr>models.<wbr>list</code></span></summary>
<p>Run <strong>List models available to this Azure resource</strong> with Azure OpenAI. This is a read command; it does not grant access beyond the connected account&#39;s permissions.</p>

<p>This is a read operation. Provider rate limits and data access rules still apply.</p>

<p><a href="https://www.tooluse.net/tools/integrations/azure-openai?command=azure-openai.models.list">Connect and use</a></p>

<p><strong>Parameters</strong></p>

<p>Requiredness is local to the containing object. A required child is required only when its parent object or array item is provided. Alternative schemas are identified in the Constraints column; the linked full JSON Schema is authoritative.</p>

<div class="nb-table-scroll"><table>
<thead>
<tr>
<th>Parameter</th>
<th>Type</th>
<th>Required in parent</th>
<th>Description</th>
<th>Constraints</th>
</tr>
</thead>
<tbody><tr>
<td><code>query</code></td>
<td>object</td>
<td>Required</td>
<td></td>
<td><code>additionalProperties: false</code></td>
</tr>
<tr>
<td><code>query.api-version</code></td>
<td>string</td>
<td>Required</td>
<td></td>
<td><code>enum: [&quot;2024-10-21&quot;]</code></td>
</tr>
</tbody></table>
</div>
<p><strong>Full JSON Schema</strong></p>

<p><a href="https://api.tooluse.net/v1/integrations/azure-openai/commands/azure-openai.models.list">Full JSON Schema</a></p>

<p><strong>Example input</strong></p>

<pre><code class="language-json">{
  "query": {
    "api-version": "2024-10-21"
  }
}
</code></pre>

<p>At most 1 provider requests per invocation; timeout: 30 seconds.</p>

</details>
<details id="command-azure-openai-chat-create">
<summary><span>Create one bounded non-streaming chat response — <code>azure-<wbr>openai.<wbr>chat.<wbr>create</code></span></summary>
<p>Run <strong>Create one bounded non-streaming chat response</strong> with Azure OpenAI. This command can change provider data. Review the target and input before invoking it.</p>

<p>This operation can change provider data. Review the input before allowing an agent to call it.</p>

<p><a href="https://www.tooluse.net/tools/integrations/azure-openai?command=azure-openai.chat.create">Connect and use</a></p>

<p><strong>Parameters</strong></p>

<p>Requiredness is local to the containing object. A required child is required only when its parent object or array item is provided. Alternative schemas are identified in the Constraints column; the linked full JSON Schema is authoritative.</p>

<div class="nb-table-scroll"><table>
<thead>
<tr>
<th>Parameter</th>
<th>Type</th>
<th>Required in parent</th>
<th>Description</th>
<th>Constraints</th>
</tr>
</thead>
<tbody><tr>
<td><code>path</code></td>
<td>object</td>
<td>Required</td>
<td></td>
<td><code>additionalProperties: false</code></td>
</tr>
<tr>
<td><code>path.deploymentId</code></td>
<td>string</td>
<td>Required</td>
<td></td>
<td><code>minLength: 1</code></td>
</tr>
<tr>
<td><code>query</code></td>
<td>object</td>
<td>Required</td>
<td></td>
<td><code>additionalProperties: false</code></td>
</tr>
<tr>
<td><code>query.api-version</code></td>
<td>string</td>
<td>Required</td>
<td></td>
<td><code>enum: [&quot;2024-10-21&quot;]</code></td>
</tr>
<tr>
<td><code>body</code></td>
<td>object</td>
<td>Required</td>
<td></td>
<td><code>additionalProperties: false</code></td>
</tr>
<tr>
<td><code>body.messages</code></td>
<td>array</td>
<td>Required</td>
<td></td>
<td><code>minItems: 1; maxItems: 20</code></td>
</tr>
<tr>
<td><code>body.messages[]</code></td>
<td>object</td>
<td>Optional</td>
<td></td>
<td><code>additionalProperties: false</code></td>
</tr>
<tr>
<td><code>body.messages[].role</code></td>
<td>string</td>
<td>Required</td>
<td></td>
<td><code>enum: [&quot;system&quot;,&quot;user&quot;,&quot;assistant&quot;]</code></td>
</tr>
<tr>
<td><code>body.messages[].content</code></td>
<td>string</td>
<td>Required</td>
<td></td>
<td><code>minLength: 1; maxLength: 8000</code></td>
</tr>
<tr>
<td><code>body.max_tokens</code></td>
<td>integer</td>
<td>Required</td>
<td></td>
<td><code>minimum: 1; maximum: 1000</code></td>
</tr>
<tr>
<td><code>body.stream</code></td>
<td>—</td>
<td>Required</td>
<td></td>
<td><code>const: false</code></td>
</tr>
<tr>
<td><code>body.temperature</code></td>
<td>number</td>
<td>Optional</td>
<td></td>
<td><code>minimum: 0; maximum: 2</code></td>
</tr>
</tbody></table>
</div>
<p><strong>Full JSON Schema</strong></p>

<p><a href="https://api.tooluse.net/v1/integrations/azure-openai/commands/azure-openai.chat.create">Full JSON Schema</a></p>

<p><strong>Example input</strong></p>

<pre><code class="language-json">{
  "path": {
    "deploymentId": "example"
  },
  "query": {
    "api-version": "2024-10-21"
  },
  "body": {
    "messages": [
      {
        "role": "user",
        "content": "Hello"
      }
    ],
    "max_tokens": 100,
    "stream": false
  }
}
</code></pre>

<p>At most 1 provider requests per invocation; timeout: 30 seconds.</p>

</details>
<details id="command-azure-openai-embeddings-create">
<summary><span>Embed one explicit text input — <code>azure-<wbr>openai.<wbr>embeddings.<wbr>create</code></span></summary>
<p>Run <strong>Embed one explicit text input</strong> with Azure OpenAI. This command can change provider data. Review the target and input before invoking it.</p>

<p>This operation can change provider data. Review the input before allowing an agent to call it.</p>

<p><a href="https://www.tooluse.net/tools/integrations/azure-openai?command=azure-openai.embeddings.create">Connect and use</a></p>

<p><strong>Parameters</strong></p>

<p>Requiredness is local to the containing object. A required child is required only when its parent object or array item is provided. Alternative schemas are identified in the Constraints column; the linked full JSON Schema is authoritative.</p>

<div class="nb-table-scroll"><table>
<thead>
<tr>
<th>Parameter</th>
<th>Type</th>
<th>Required in parent</th>
<th>Description</th>
<th>Constraints</th>
</tr>
</thead>
<tbody><tr>
<td><code>path</code></td>
<td>object</td>
<td>Required</td>
<td></td>
<td><code>additionalProperties: false</code></td>
</tr>
<tr>
<td><code>path.deploymentId</code></td>
<td>string</td>
<td>Required</td>
<td></td>
<td><code>minLength: 1</code></td>
</tr>
<tr>
<td><code>query</code></td>
<td>object</td>
<td>Required</td>
<td></td>
<td><code>additionalProperties: false</code></td>
</tr>
<tr>
<td><code>query.api-version</code></td>
<td>string</td>
<td>Required</td>
<td></td>
<td><code>enum: [&quot;2024-10-21&quot;]</code></td>
</tr>
<tr>
<td><code>body</code></td>
<td>object</td>
<td>Required</td>
<td></td>
<td><code>additionalProperties: false</code></td>
</tr>
<tr>
<td><code>body.input</code></td>
<td>string</td>
<td>Required</td>
<td></td>
<td><code>minLength: 1; maxLength: 8000</code></td>
</tr>
</tbody></table>
</div>
<p><strong>Full JSON Schema</strong></p>

<p><a href="https://api.tooluse.net/v1/integrations/azure-openai/commands/azure-openai.embeddings.create">Full JSON Schema</a></p>

<p><strong>Example input</strong></p>

<pre><code class="language-json">{
  "path": {
    "deploymentId": "example"
  },
  "query": {
    "api-version": "2024-10-21"
  },
  "body": {
    "input": "Example text"
  }
}
</code></pre>

<p>At most 1 provider requests per invocation; timeout: 30 seconds.</p>

</details>
<h2>Errors</h2>

<p>Check the key&#39;s expiry, revocation, workspace and allowed operations.</p>

<p>Reuse the same idempotency key only for the same request.</p>

<p>The connection or provider limit has been reached. Check the receipt before retrying.</p>

<h2>Test the connection</h2>

<p>Open the integration&#39;s commands below to inspect parameters and examples. Start with a read command for an account or resource you can access. Refresh your client&#39;s tool list, use the returned tool name, and check the execution result and usage. Provider charges and permissions still apply. Never test a write against production data without reviewing it.</p>

<p><a href="https://learn.microsoft.com/en-us/azure/ai-services/openai/reference">Official provider documentation</a></p>

Source: https://help.tooluse.net/integrations/azure-openai/index.mdx