> For the complete documentation index, see [llms.txt](https://documentation.astera.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://documentation.astera.com/astera-studio/workspace/ai-skills.md).

# AI Skills

A Skill is a reusable instruction set that teaches an AI agent how to perform a specific task — for example, validate a customer dataset, summarise a weekly sales report, or publish a dashboard to the marketing team. Skills bundle clear instructions with a curated list of MCP tools, so agents have both the what to do and the what to use in one place. 

In this article, we will learn how Skills are structured, where they live in a workspace, and how to create, edit, and import them — including the AI-assisted drafting modes that let you generate a Skill from a one-line intent or from an existing chat conversation. 

## New Skill

To create a new skill, click **Add Item** at the top of the Workspace Panel and select **Skill** under the Artifacts section.

<figure><img src="/files/P5Mltv3g2VRF3S3EBbBC" alt="" width="191"><figcaption></figcaption></figure>

This opens the **Teach the AI a new skill** modal, where you can choose how to add a new skill to your workspace.

<figure><img src="/files/nSiWpPiNL0l8OAHWhnVa" alt=""><figcaption></figcaption></figure>

### Draft with AI &#x20;

Type a one-line description of what the Skill should do into the input field and click **Draft**. Astera Studio generates a full Skill including metadata, instructions, and a suggested set of bound tools. You can select the model to use for drafting from the **Model** dropdown. Best when you know what you need but want to skip the blank page.

### Distill from active chat

Only enabled when a chat is open. Astera Studio reviews the conversation, and the tool calls in it, then distils them into a reusable Skill. Best after you have walked an agent through a task once and want to capture it.&#x20;

<figure><img src="/files/Jx7Rcf60QTKDGi1Xeluu" alt="" width="563"><figcaption></figcaption></figure>

### Start blank&#x20;

Opens an empty editor with no AI assistance. Best when you want full control over every line.&#x20;

![](/files/kss4AdjjFmfne1fP02VE)

#### Editor&#x20;

This is where you write the Skill itself. The tab has two halves:&#x20;

* **Front Matter (left side) —** contains the Alias, Short Description, Category, and Tags fields. These make up the skill's front matter, which the agent references when determining which skill to apply to a given query. Hence it is necessary that these fields are filled out accurately. Vague or missing metadata will cause the agent to misroute or overlook the skill entirely.&#x20;
  * **Alias** - A short, machine-readable identifier. Lowercase letters and hyphens only. Required when you create a Skill; cannot be changed afterwards.
  * **Short Description** - A one-sentence summary that helps the AI decide when the Skill applies. Keep it under 255 characters.
  * **Category** - An optional grouping label so related Skills cluster together in the Skills list.
  * **Tags** - Free-form labels for search.
* **Skill Content (right side)** — the main editor where you write the instructions the agent will follow when this skill is applied. Content is written in Markdown. You can author it manually or use **Draft with AI** to generate a starting draft.&#x20;

#### Bound Tools

This is where you tell the Skill which MCP tools it is allowed to use. Without bound tools, a Skill can only reason; with bound tools, it can take action.&#x20;

1. Click the Bound Tools tab in the editor.&#x20;

<figure><img src="/files/KNwKVrve7ew10jBS67ld" alt=""><figcaption></figcaption></figure>

2. Browse the tool tree on the left. Tools are grouped by category, and each category can be expanded or collapsed.&#x20;
3. Search by tool name or category using the search field at the top of the tree.&#x20;
4. Tick the checkbox next to any tool you want this Skill to use. The right-hand panel shows the tool's parameters and description.&#x20;

<figure><img src="/files/DoP0SClobNdg5SdlRain" alt=""><figcaption></figcaption></figure>

5. Click Save in the toolbar when you are done. The skill will be added to your workspace ready to be used by the agent.

## From Markdown

If you already have skill content in markdown form. Perhaps exported from another workspace or written outside Astera Studio, you can import it without retyping.

1. Click ***Add Item> From Markdown.***&#x20;

<figure><img src="/files/7LfU0dylS4gkyMEY3t9q" alt="" width="194"><figcaption></figcaption></figure>

2. Pick a .md file from cloud storage.&#x20;

<img src="/files/F9PK0SEweuZPCvIhtgTo" alt="" width="563">

3. Astera Studio parses the file's front-matter and body, opens the Skill Editor, and lets you review and save.&#x20;

<figure><img src="/files/U9j1Mb8IdrLlBHpQqequ" alt=""><figcaption></figcaption></figure>

## From Markdown Folder

1. Click ***Add Item > From Markdown Folder.***&#x20;

<figure><img src="/files/Pd8i4HL5EEQtYgjiQLsA" alt="" width="194"><figcaption></figcaption></figure>

2. Pick a folder. Astera Studio scans for .md files (including nested folders) and imports each one as a Skill.&#x20;

<img src="/files/dhHNGV49X2DaSKP1P2jC" alt="" width="563">

3. All the skills get imported and added to your workspace. You can open each one to verify or edit further.&#x20;


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://documentation.astera.com/astera-studio/workspace/ai-skills.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.