> For the complete documentation index, see [llms.txt](https://developers.make.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developers.make.com/custom-apps-documentation/get-started/make-apps-editor/ai-coding-agent.md).

# AI coding agent

You can build or modify custom apps with an AI coding agent, such as Claude Code or Cursor. The agent uses the following tools:

* The [Make custom apps skill](https://github.com/integromat/make-apps-skill) for instructions on building custom apps
* The [Make TypeScript SDK](https://developers.make.com/make-sdk/make-typescript-sdk) for creating, updating, and reading app definitions
* [Make Skills](https://github.com/integromat/make-skills) and [Make MCP server](https://developers.make.com/mcp-server) for building and running test scenarios (optional)

### What you'll need

Before you start building, have these items ready:

* An AI coding agent that reads project files and runs local commands.
* A Make account with [permission to manage custom apps](https://developers.make.com/white-label-documentation/install-and-configure-apps/manage-custom-apps).
* A Make team to test in, separate from your production scenarios and connections.
* Node.js 20 or later.
* The third-party service's API documentation or OpenAPI file.
* A test account for the third-party service. If available, use a sandbox account.

You'll also create a Make API key in the first step.

### App development process

Follow these steps to create a custom app using an AI coding agent.

#### Step 1. Set up your agent

Install the Make custom apps skill, create a Make API key, and give the agent access to the Make developer documentation.

See [Set up your agent](/custom-apps-documentation/get-started/make-apps-editor/ai-coding-agent/set-up-your-agent.md).

#### Step 2. Build and upload your app

Brief the agent, approve its design, review the generated files, and approve the upload of a private app to Make for testing.

See [Build and upload your app](/custom-apps-documentation/get-started/make-apps-editor/ai-coding-agent/build-and-upload-your-app.md).

#### Step 3. Connect and test your app

Create a connection to the third-party service, run the app's modules in a test scenario, and check the output.

See [Connect and test your app](/custom-apps-documentation/get-started/make-apps-editor/ai-coding-agent/connect-and-test-your-app.md). 

{% hint style="info" %}
**Keep in mind while building:**

* A skill gives your agent instructions. It doesn't give your agent credentials or access to your Make account.
* The Make custom apps skill installs separately from Make Skills.
* Keep secrets in environment variables or a secret store, not in agent prompts or source control.
  {% endhint %}

{% hint style="warning" %}
The Make Apps Editor extension for VS Code is deprecated. Follow the workflow in this section instead. You can still use VS Code as the editor for your AI coding agent.
{% endhint %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://developers.make.com/custom-apps-documentation/get-started/make-apps-editor/ai-coding-agent.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.