> For the complete documentation index, see [llms.txt](https://docs.appmixer.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.appmixer.com/build-connectors-with-a-coding-agent/copilot-vs-skills.md).

# AI Copilot vs. Appmixer Skills

Appmixer has two AI features with little in common beyond the name, and a third product whose name collides with one of them. This page draws the line once.

## Two tools, two jobs

|                | AI Copilot                                                            | Appmixer Skills                                            |
| -------------- | --------------------------------------------------------------------- | ---------------------------------------------------------- |
| **Builds**     | flows                                                                 | connectors                                                 |
| **Runs in**    | the Designer, as a system plugin of the engine                        | your coding agent, on your machine                         |
| **Used by**    | anyone building automations on an instance                            | a developer writing a connector                            |
| **Works from** | components already on the instance                                    | source files in your workspace, through the `appmixer` CLI |
| **Produces**   | a flow on the canvas, or a draft of a running flow                    | a connector published to an instance                       |
| **Set up by**  | an administrator — `SYSTEM_PLUGINS=copilot` plus `COPILOT_*` settings | you — a plugin or skills install in your agent             |
| **Model**      | whatever the instance is configured with                              | whatever your coding agent runs                            |

Neither does the other's job. Copilot's tools read apps, components and variables and save flows; none writes a component. The skills create no flows for end users — only the E2E test flows for the connector they build.

For Copilot, see [AI Copilot](https://github.com/Appmixer-ai/appmixer-docs-gitbook/tree/6.5/getting-started/ai-copilot.md). For the skills, start at the [Overview](/build-connectors-with-a-coding-agent/agent-connectors.md).

## Where they meet

**The connector you publish becomes something Copilot can build with**, with no extra step. Copilot finds components the way the platform does, filtered by each user's component ACL.

What Copilot knows about your connector is what the manifests say. It reads `aiDescription`, falling back to `description` where it is missing — so `aiDescription` is the one field written for Copilot rather than for people, and nothing checks it ([Scope and Research](/build-connectors-with-a-coding-agent/research.md)).

If Copilot never uses one of your components, or uses the wrong one, check `aiDescription`, then the user's component ACL.

## The name collision: GitHub Copilot

**GitHub Copilot is a coding agent made by GitHub.** It can run Appmixer Skills, installed like any other agent:

```bash
npx skills add Appmixer-ai/appmixer-skills
```

It has nothing to do with Appmixer's AI Copilot. In this guide, "AI Copilot" means the Designer feature; the coding agent is always "GitHub Copilot".

## And `appmixer ai`

The CLI still ships `appmixer ai connector` and `appmixer ai component`, an earlier generator that wrote a connector from API documentation or an OpenAPI URL. The CLI's README marks them **deprecated** — superseded by the skills and due to be removed. Do not start new work on them.

## Which one you want

* **"I want a flow that does X, with services already on my instance."** AI Copilot.
* **"The service I need is not on my instance."** Appmixer Skills and this guide. Once published, AI Copilot can use your connector too.
* **"AI Copilot will not use my connector."** Check `aiDescription`, then the user's component ACL.


---

# 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://docs.appmixer.com/build-connectors-with-a-coding-agent/copilot-vs-skills.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.