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

# Skills

Save your recurring instructions as reusable commands in the Dastra AI chat

A **skill** is an instruction you save once and then recall with a command, by typing `/` in the conversation. Think of it as a prompt template shared across the organization.

Skills are managed in **Automation** > **Dastra AI** > **Skills**.

<figure><img src="https://2697025545-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-LvBxs22wUMicv9uWp6C-1972196547%2Fuploads%2Fgit-blob-33ff4da53beecd262a0947c9629e1971153ddf61%2Fai-agents-competences.png?alt=media" alt="List of Dastra AI skills"><figcaption><p>The Skills tab, with Dastra's built-in skills</p></figcaption></figure>

***

### Built-in skills

Dastra ships three skills, available in every workspace and not editable:

| Command                 | What it does                                                                                                                                                                                   |
| ----------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `/design_agent`         | Guides you through designing a custom agent (persona, tools, rules) and saves it once you validate. This is what the **Generate with Dastra AI** option of the agent creation screen triggers. |
| `/dpo_opinion`          | Turns the current discussion, or the record you are looking at, into a formal, archivable DPO opinion.                                                                                         |
| `/design_workflow_rule` | Guides you through building a workflow rule (trigger, conditions, actions) based on your workspace, and enables it once you validate.                                                          |

### Creating your own skill

Click **Create a skill**, then fill in:

* **Name**: the label shown in the command list (120 characters maximum).
* **Command**: the text typed after the `/`. Lowercase letters, digits and underscores, starting with a letter, and **unique within the organization**: for example `steering_committee_summary`.
* **Description**: one sentence explaining when to use it, shown next to the command in the chat (1000 characters maximum).
* **Prompt**: the instruction sent to the assistant, up to 10,000 characters.
* **Share with the organization**: when it is on, every member of your organization sees the skill, in all the workspaces. Otherwise only you can use it. The list shows each skill as **Shared** or **Private**.

### Using a skill

In a Dastra AI conversation, type `/` or click **Skills**: the list of available skills appears, with the **Built-in skills** and your **Custom skills**. Pick one, complete your message if needed, then send. A message applies at most five skills.

{% hint style="info" %}
A skill applies to the current conversation, with the agent and the mode (Ask or Edit) currently selected. A skill that needs to modify records therefore requires Edit mode.
{% endhint %}

Creating, updating or deleting a skill requires the write level of the **Custom AI agents** permission, but not the custom agents plan option. Using a skill only requires access to the assistant. The **Skills** list of the menu appears with the other custom agent entries. From the **Skills** submenu of the message box, you can also add a skill (**Add a skill**), or edit and delete your own, without leaving the chat.

***

### Skill or agent?

Both let you capitalise on an instruction, but not at the same moment:

* A **skill** is a one-off. You trigger it in the middle of a conversation, on whatever context is at hand.
* An **agent** is permanent. It has its own identity, its own tools, and where relevant its own schedule.

If you keep retyping the same request, create a skill. If you want a specialised counterpart or an automated task, create an [agent](https://doc.dastra.eu/en/dastra-ai/ai-assistant/agents.md).


---

# 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://doc.dastra.eu/en/dastra-ai/ai-assistant/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.
