> ## Documentation Index
> Fetch the complete documentation index at: https://docs.b-bot.space/llms.txt
> Use this file to discover all available pages before exploring further.

# DeepAgents

> Understand how DeepAgents use tasks, workspace files, tools, and human review for multi-step work.

# DeepAgents

DeepAgents is B-Bot Hub's mode for work that benefits from planning and intermediate artifacts. A DeepAgent can use a task list and workspace files while working through a multi-step request. The exact tools and actions available depend on the Agent, session, and environment.

## Core Concepts

<CardGroup cols={2}>
  <Card title="Tasks" icon="list-check">
    A visible plan can divide a goal into smaller steps and show progress.
  </Card>

  <Card title="Workspace Files" icon="folder-open">
    Inputs, notes, code, data, and outputs can be organized in the workspace.
  </Card>

  <Card title="Tools" icon="toolbox">
    Configured tools may let the DeepAgent retrieve information, process files, run code, or interact with connected services.
  </Card>

  <Card title="Human Review" icon="hand">
    Human-in-the-loop controls can require review before selected actions.
  </Card>
</CardGroup>

## Conceptual Workflow

```text theme={null}
Goal
  -> clarify requirements
  -> create or update tasks
  -> use permitted tools and workspace files
  -> check intermediate results
  -> request approval where required
  -> present outputs and unresolved issues
```

A task marked complete only indicates workflow state. It does not prove that the result is correct.

## When to Use DeepAgents

DeepAgents can be useful for:

* Projects with several dependent steps
* Work that produces or revises multiple files
* Analysis that needs visible assumptions and intermediate checks
* Long requests that benefit from a maintained task list
* Workflows where people need to review proposed actions

Standard chat may be simpler for a direct question, a short rewrite, or another task that does not need a workspace.

## Giving a DeepAgent a Good Request

Include:

* The required outcome and audience
* Inputs and source-of-truth material
* Constraints, exclusions, and quality criteria
* Allowed tools and actions
* Required file names or output formats
* Actions that require approval
* A definition of done

Example:

```text theme={null}
Analyze the attached CSV and prepare a reviewable report.

Create tasks for data inspection, cleaning, analysis, validation, and reporting.
Keep the source file unchanged.
Record every transformation in a separate file.
Ask for approval before running code that writes new output files.
Do not send data to external services.
The work is complete when totals are reconciled and the report lists assumptions.
```

## Human-in-the-Loop

Use human review for actions that are sensitive, irreversible, externally visible, or costly. Examples include:

* Sending messages or publishing content
* Changing records or permissions
* Running unfamiliar code
* Using private data with an external service
* Triggering purchases, payments, or deployments

Human-in-the-loop reduces risk but does not make an action safe by itself. The reviewer still needs enough context to understand the proposal.

## Workspace Practices

* Keep original inputs separate from generated files.
* Use descriptive file names.
* Record assumptions and decisions.
* Review generated code before execution.
* Download or retain important outputs according to your data policy.
* Verify workspace behavior and retention for the specific workflow.

<Warning>
  Grant only the tools and permissions needed for the task. A system prompt is not an access-control boundary.
</Warning>

## Related Guides

<CardGroup cols={2}>
  <Card title="DeepAgents Workspace" icon="folder-tree" href="/essentials/deepagents-workspace">
    Learn how to enable and use workspace features.
  </Card>

  <Card title="Tasks" icon="calendar-check" href="/essentials/tasks">
    Learn about scheduled and on-demand tasks.
  </Card>

  <Card title="Data Analysis" icon="chart-line" href="/concepts/use_cases/data-analysis">
    Apply DeepAgents to a reviewable analysis workflow.
  </Card>

  <Card title="Content Creation" icon="pen-to-square" href="/concepts/use_cases/content-creation">
    Organize a multi-stage editorial workflow.
  </Card>
</CardGroup>