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

# Skills

AI agent skills are multi-step jobs your agent runs end to end, like testing its answers on past tickets or finding ticket trends. Learn how to run one or make your own.

A skill is a job your teammate knows how to do end to end: a multi-step job with a set process and a saved result. Your teammate testing its answers against hundreds of your past tickets is a skill. Reading a month of conversations and writing up what customers keep asking about is a skill.

A skill can also act on what it finds, so the work gets finished rather than handed back to you.

eesel ships a set of skills, and on newer workspaces your team can add its own. See [Create a custom skill](#create-a-custom-skill).

## Quick start

{% stepper %}
{% step %}

### Ask your agent to run a skill

Open the chat in your dashboard and ask for the skill by name, or describe the job you want done. Your agent picks the skill, gathers what it needs and reports each step in the chat as it goes.

<figure><img src="https://3732419023-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F2xXO0947TYoPIhGoBgSE%2Fuploads%2Fgit-blob-6cad8c003908b0f9c07104b38a811d245781fd45%2Fskills-run-chat.png?alt=media" alt="A dashboard chat where the user asks the agent to run the Helpdesk simulation skill, and the agent reports each step as it samples 20 resolved tickets and starts the runs"><figcaption><p>The Helpdesk simulation skill starting from one line of chat.</p></figcaption></figure>

If the skill needs an app you have not connected, it tells you and walks you through connecting it.
{% endstep %}

{% step %}

### Read the result

When it is done, the result comes back in the same chat: the headline finding first, then the detail.

<figure><img src="https://3732419023-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F2xXO0947TYoPIhGoBgSE%2Fuploads%2Fgit-blob-bc79839d5b1f7aee7ddaa83bb450f92ffaea441c%2Fskills-run-results.png?alt=media" alt="The chat showing the simulation result: 17 of 20 tickets matched the team&#x27;s reply quality, with a score for each ticket theme"><figcaption><p>The headline score, then the same score split by theme.</p></figcaption></figure>

It ends with what it suggests you do next, a link to the full output, and a note of anything it sent on your behalf.

<figure><img src="https://3732419023-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F2xXO0947TYoPIhGoBgSE%2Fuploads%2Fgit-blob-c4129260e26c9f6b5bff1f3be9649dbccdfffd83%2Fskills-run-fixes.png?alt=media" alt="The same chat scrolled down, listing five ranked fixes, a link to the full simulation report and a note that the summary was emailed"><figcaption><p>Five ranked fixes, the report link, and the email it sent.</p></figcaption></figure>
{% endstep %}

{% step %}

### Open the full report

Click the link, or the file under **Files** in the left sidebar. The report opens as a document beside the chat, so you can read it and act on it in the same place.

<figure><img src="https://3732419023-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F2xXO0947TYoPIhGoBgSE%2Fuploads%2Fgit-blob-89d8bd80a6d67fe3bc81c66743e09d7c94e2ef71%2Fskills-run-report.png?alt=media" alt="The Simulation report file open in the dashboard next to the chat, showing the headline finding, four result tiles and the results by theme"><figcaption><p>The saved report, with the chat still beside it.</p></figcaption></figure>

A skill can also run on its own. Attach it to an [automation](https://docs.eesel.ai/automations-and-schedules/automations-and-schedules.md) and it runs on a schedule, for example a report every Monday morning.
{% endstep %}

{% step %}

### Or start from the Skills page

Open **Skills** in the left sidebar. It has two tabs: **eesel skills**, the default skills that come with every agent, and **Your skills**, the custom skills your team creates.

eesel's skills are shown as sample reports, made with fictional data so you can see what each one produces. Open one to see the sample, or ask for the real one, built from your own data, in a new chat. They are requests, not things you install. If one needs an app you have not connected, your agent walks you through connecting it.

<figure><img src="https://3732419023-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F2xXO0947TYoPIhGoBgSE%2Fuploads%2Fgit-blob-e1ab270738c38d65e1fad7ae3d9e9df14796be71%2Fskills-page-custom-and-default.png?alt=media" alt="The Skills page with two custom skills, reply-checker and weekly-refund-audit, above eesel&#x27;s sample reports"><figcaption><p>Custom skills on top, eesel's default skills below.</p></figcaption></figure>
{% endstep %}
{% endstepper %}

## Create a custom skill

When there is a job you want done the same way every time, like a weekly refund audit, ask your agent to save it as a skill, or start one from the Skills page. Your agent writes the steps into a `SKILL.md` file, with any scripts the job needs beside it. Next time, ask for the job by name and your agent follows the saved steps.

Custom skills are saved with your agent's files, and you can edit one any time: open it and change its `SKILL.md`. To change a script it runs, ask your agent.

If your custom skills don't show under **Files** in the sidebar, turn on **Show hidden files** with the eye icon on the Files row.

<figure><img src="https://3732419023-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F2xXO0947TYoPIhGoBgSE%2Fuploads%2Fgit-blob-076848158c466173f057265757dd721e1ebcfaec%2Fskills-files-sidebar.png?alt=media" alt="The Files section of the sidebar with Show hidden files turned on, showing the .skills folder open: the Reply checker folder holds Check reply.py and Reply checker skill, its SKILL.md, beside a Weekly refund audit folder" width="258"><figcaption><p>Custom skills in Files, with hidden files shown.</p></figcaption></figure>

<figure><img src="https://3732419023-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F2xXO0947TYoPIhGoBgSE%2Fuploads%2Fgit-blob-de90be8f905d5c841acb588ad040a08909c1ea64%2Fskills-custom-skill-md.png?alt=media" alt="A custom skill&#x27;s SKILL.md open in the dashboard, showing its steps and how to run its script"><figcaption><p>A custom skill's SKILL.md, open to edit.</p></figcaption></figure>

A skill is for a job with its own result that someone asks for. How to handle a kind of conversation, like damaged-item tickets, belongs in your [instructions or memory](https://docs.eesel.ai/instructions-and-memory/instructions-and-memory.md), because those apply every time a conversation fits.

Custom skills are for your team: your agent uses them when it works for you, not in conversations with your customers. eesel's default skills cannot be edited. To change how one behaves, tell your agent and it adds the change to your instructions.

## The sample reports on the Skills page

Each card is named after what it produces, and there is nothing to find or switch on for it. You can ask for one by its name in chat, for example "send me the weekly digest every Friday", and your agent does what the card's button does, fitted to what you asked for.

| Card                           | What it covers                                                                                                                                                                                                                                                                     |
| ------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Path to automation**         | Your helpdesk tickets: what kinds you get, how much it can handle today, counted from real tickets, what each connection or instruction unlocks, what stays with your team, and the steps in order                                                                                 |
| **Simulation on past tickets** | Tickets your team recently resolved, spread across your main topics, and your live chats if you have them: it drafts its own reply to each, puts it next to what your team sent, groups the results by theme and says what to fix before going live. It sends nothing to customers |
| **What customers are asking**  | The last 30 days of tickets and chats: what customers ask most, what is rising, which questions your help center misses or answers badly, and the missing articles drafted in full                                                                                                 |
| **Weekly digest**              | Last week's activity: conversations handled and how, by topic and channel, approvals waiting, what it skipped and why, and what to decide this week. The card also sets it up to send every Monday                                                                                 |
| **Slide deck**                 | A designed deck for any audience. The sample is the case for eesel, made for your leadership from your own tickets: your volume, what eesel could take on, one real ticket start to finish, and what going live needs                                                              |
| **Blog post**                  | A researched, publish-ready blog post in your brand's voice, on a topic your customers search for. See [Content Writing](https://docs.eesel.ai/use-cases/content-writing.md)                                                                                                       |

## What ships out of the box

| Skill                                                                                         | What it does                                                                                                                                                                            |
| --------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Helpdesk simulation**                                                                       | Runs your agent against real past tickets or generated test cases, scores each answer, and suggests instruction changes                                                                 |
| **Analyze and improve replies**                                                               | Looks at what your team rejected or edited, finds the pattern, and suggests fixes                                                                                                       |
| **Review agent setup**                                                                        | Audits your agent's configuration and performance, and tells you what to change                                                                                                         |
| **Analyze ticket trends**                                                                     | Recurring topics, sentiment, volume and resolution patterns across your conversations                                                                                                   |
| **Update knowledge base**                                                                     | Finds the questions your help center does not answer, and drafts the missing articles. Needs switching on for your workspace first, so [contact us](https://eesel.ai/contact) to use it |
| **Write a blog post**                                                                         | Researches and writes a full post with images. See [Content Writing](https://docs.eesel.ai/use-cases/content-writing.md)                                                                |
| **Translate content**                                                                         | Translates content while keeping formatting and meaning intact                                                                                                                          |
| **Draft an internal note**                                                                    | Writes a reply as an internal note for your team to review                                                                                                                              |
| **Send a public reply**                                                                       | Sends a reply to the customer                                                                                                                                                           |
| [**Look up Calendly meetings**](https://docs.eesel.ai/integrations/communication/calendly.md) | Reads your Calendly bookings: who a meeting is with, who is hosting, when, and whether it was canceled. Read-only                                                                       |
| **Voice call transcriber**                                                                    | Transcribes call recordings                                                                                                                                                             |
| **Commit and push to GitHub**                                                                 | Commits and pushes changes to a repository                                                                                                                                              |

## FAQ

<details>

<summary>What is a skill in eesel?</summary>

A skill is a multi-step job your agent knows how to do end to end, with a set process and a saved result. Testing its answers against your past tickets is a skill, and so is writing up what customers keep asking about.

</details>

<details>

<summary>How do I run a skill?</summary>

Ask for it in the dashboard chat, by name or by describing the job. Or open **Skills** in the left sidebar and start one of the sample reports. Your agent reports each step in the chat and saves the full result under **Files**.

</details>

<details>

<summary>Can a skill run on a schedule?</summary>

Yes. Attach it to an [automation](https://docs.eesel.ai/automations-and-schedules/automations-and-schedules.md) and it runs on a schedule or a trigger, for example a report every Monday morning.

</details>

<details>

<summary>My Skills page shows a list of skills, not sample reports</summary>

That is the earlier Skills page, which some workspaces still have: a list where each row is a skill. Open one, click **Install skill**, then **Try in chat** to start it. A skill marked **Contact us** needs switching on for your workspace first: [contact us](https://eesel.ai/contact) or ask in the in-app support chat under **Help and support**.

</details>

<details>

<summary>Is there a community marketplace?</summary>

Not yet. eesel's skills are built by eesel, and your team can [make its own](#create-a-custom-skill) for itself. You use either by asking your agent for it in chat. From a terminal, `eesel chat "analyze our ticket trends for last month"` does the same through the [CLI](https://docs.eesel.ai/apis-and-developer-resources/cli.md). A community marketplace where teams publish skills for each other is something we are working towards.

</details>

<details>

<summary>Can a skill work across more than one app?</summary>

Yes. A skill can use anything your agent is connected to, in one run: read your helpdesk, check Shopify, and post the result to Slack. A skill can also reach your own systems through [Your APIs](https://docs.eesel.ai/apis-and-developer-resources/network-access.md).

</details>

<details>

<summary>Where do I see what a skill produced?</summary>

Every run shows up in **Activity** with its output attached. Outputs sent to a file live under **Files**.

</details>

<details>

<summary>Can a skill have its own instructions?</summary>

Yes. A skill can carry saved instructions that apply only when it runs, so a blog style guide does not sit in the instructions your support replies read. Your agent's core [instructions](https://docs.eesel.ai/instructions-and-memory/instructions-and-memory.md) still apply underneath.

</details>

## Related

* [Automations and Schedules](https://docs.eesel.ai/automations-and-schedules/automations-and-schedules.md) to run a skill on a schedule or a trigger
* [Instructions and Memory](https://docs.eesel.ai/instructions-and-memory/instructions-and-memory.md) for how skills feed back into how your agent behaves
* [Reports](https://docs.eesel.ai/reports/reports.md) for the report skills and the Reports page
* [Use eesel from Claude and other AI assistants](https://docs.eesel.ai/apis-and-developer-resources/use-eesel-from-claude-and-other-assistants.md) and [Use eesel from ChatGPT](https://docs.eesel.ai/apis-and-developer-resources/use-eesel-from-chatgpt.md) to ask your agent for a skill from a chat assistant


---

# 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.eesel.ai/skills/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.
