> For the complete documentation index, see [llms.txt](https://docs.ebrain.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.ebrain.ai/agents-autonomous/overview.md).
# What are Agents?
An Agent is a chat with Tim that you have saved so it can run on its own. You describe, in plain language, the one job you want done each time it runs, and eBrain carries it out automatically and delivers the result to you.
You find your Agents on the **Agents** page, summed up by its subtitle: **"Run chats on a schedule, or whenever you need them."**
## Agents vs. chatting with Tim
The difference comes down to one thing: whether you are present.
When you chat with Tim, it is a live, back-and-forth conversation. You are in the room, Tim can ask you questions, and he asks for your approval before any write action like sending an email or creating a calendar event.
An Agent runs **unattended**. There is no one there to answer a question or approve a step partway through. Because of that, an Agent works in **read-only** mode: it gathers and analyzes your information (mail, calendar, tasks, and saved data) and produces an output, such as a briefing, a summary, a draft, or a list of what needs your attention. It then delivers that result for you to review and act on.
{% hint style="info" %}
Agents produce drafts and summaries for you to review. They do not send emails or create calendar events on their own. Any sending or scheduling still happens when you act on the result yourself, with Tim, where the usual approval step applies.
{% endhint %}
The short version:
| | Chatting with Tim | Agents |
| --------------- | -------------------------------------- | --------------------------------------------- |
| When it happens | Now, while you are there | Later, or on a repeat, by itself |
| Style | Live, back-and-forth | Unattended, runs to completion |
| Writes | Sends and creates with your approval | Read-only; produces results for you to review |
| Good for | One-off requests, anything interactive | Recurring jobs you do not want to think about |
## When to use an Agent
Reach for an Agent whenever a useful task repeats and you would rather not remember to ask for it. Instead of opening a chat every morning to ask "what needs my attention today?", you save that request once and let it run on a schedule. Here are a few concrete examples:
{% stepper %}
{% step %}
## Daily brief
A morning summary of what needs your attention across email, tasks, and invoices, delivered every day so you start the day already informed.
{% endstep %}
{% step %}
## Weekly review
A weekly recap of progress, blockers, and what is coming up, for example every Friday afternoon.
{% endstep %}
{% step %}
## Inbox triage
Sorts and prioritizes incoming email so the important things rise to the top, running through the day.
{% endstep %}
{% step %}
## Project monitor
Keeps an eye on a specific project and surfaces overdue or stalled work, scoped to that project.
{% endstep %}
{% endstepper %}
eBrain also ships ready-made templates (such as Meeting prep and Release notes) that set up agents like these in a couple of clicks. You can adapt any of them or describe your own from scratch.
## How an Agent runs and where the result goes
Every run posts a fresh conversation into your **Chats**, tagged as an agent run so you can tell it apart from your own chats, and you get an in-app notification when a run finishes. From there you can read everything the agent produced and continue with Tim. You can optionally have results also delivered to WhatsApp or email.
An Agent can run **on a schedule** (for example every morning) or stay **on demand**, meaning it has no schedule and only runs when you press **Run now**. Either way, you stay in control: you can pause a scheduled agent, run any agent immediately, and review every past run.
{% hint style="info" %}
The easiest way to create an agent is to describe it to Tim. He asks any clarifying questions, writes the agent's instructions for you, and proposes it as a card you approve.
{% endhint %}
## Next steps
---
# 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:
```
GET https://docs.ebrain.ai/agents-autonomous/overview.md?ask=
```
The question should be specific, self-contained, and written in natural language.
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.