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

# Agents

> Agent anatomy, configuration, solo agents, skills, MCP tools, and permission modes.

An agent is a single AI worker. Agents can run inside a team or as a solo workspace.

## Anatomy

| Part                 | What it is                                                                     |
| -------------------- | ------------------------------------------------------------------------------ |
| **Name**             | Display name, such as "Schema Wrangler"                                        |
| **Specialization**   | Role and baked-in skill set from the Marketplace                               |
| **Model**            | The provider model powering the agent                                          |
| **Soul**             | Markdown prompt defining expertise and behavior                                |
| **Skills**           | Slash-command workflows from its specialization                                |
| **Tools (MCP)**      | External tools enabled per agent                                               |
| **Context settings** | How much history, task context, messages, references, and tool context it sees |
| **Permission mode**  | What it may do without asking                                                  |

## Team Agents vs Solo Agents

|                | Team agent                           | Solo agent              |
| -------------- | ------------------------------------ | ----------------------- |
| Best for       | Work that benefits from coordination | Focused one-role work   |
| Manager        | Atlas delegates and tracks tasks     | No Manager              |
| Chat           | Part of a team workspace             | Direct 1:1 conversation |
| Working folder | Team working directory               | Solo working directory  |
| Modes          | Permission mode controls autonomy    | Chat mode or Build mode |

Solo **Chat** mode is best for questions, planning, writing, research, and review. Solo **Build** mode lets the agent edit files and run commands in its folder.

## Statuses

Agents show a live status in the sidebar, canvas, and headers:

* **Active**: currently working.
* **Idle**: waiting for work.
* **Waiting**: needs approval or input.
* **Blocked**: hit an error; open chat for details.

## Configuration Panel

Open an agent's settings to tune it.

### Identity

Name, specialization badge, and model selector. Models are listed by connected provider.

### Soul

The agent's system prompt as markdown. Built-in specializations are read-only. User-made agents expose an editable section.

<Warning>
  Saved Soul edits cannot be undone automatically. Copy the current text somewhere safe before large rewrites.
</Warning>

### Context

Controls what the agent sees on each run:

* **History & Memory**: message lookback, max characters per message, and role filter.
* **Scope & Budget**: team-message lookback, reference max, and token budget.
* **Sections**: Team, Connectors, Tasks, Messages, References.
* **Prompt Analytics**: prompt size, estimated tokens, and per-section breakdown.

### Skills

Skills are packaged workflows agents can run with slash commands.

* Built-in specializations include curated skills.
* User-made packages come from [Agent Creator](/guides/agent-marketplace/agent-maker).
* Community packages come from GitHub, local folders, or CLI install commands.

Type **/** in chat to open the slash-command menu. You can also ask in plain language; slash commands are the explicit way to trigger a workflow.

Suggested workflows appear in the Marketplace for built-in specializations.

### Tools (MCP)

MCP servers connect agents to external tools and data, such as GitHub, databases, or internal APIs.

Install MCP servers once in Settings, then enable them per agent. Enable only what the role needs; fewer tools means better tool choices and a smaller blast radius.

Common setup paths:

* **Catalog**: one-click presets for common servers.
* **Manual server**: local command, remote HTTP URL, or remote SSE URL.
* **Secrets**: encrypted locally and injected at runtime.

MCP tool calls go through the same permission flow as built-in tools.

### Danger Zone

**Delete Agent** removes the agent and all of its configuration. Project files are not deleted.

## Permission Mode

Set permission mode from the shield selector:

| Mode                  | Behavior                                                  |
| --------------------- | --------------------------------------------------------- |
| **Ask first**         | Every tool use prompts you                                |
| **Auto-accept edits** | File reads/edits run automatically; commands still prompt |
| **Full auto**         | Commands, edits, and tools run without prompts            |

Details and recommendations live in [Permissions and safety](/reference/permissions-and-safety).

## Next Steps

<CardGroup cols={2}>
  <Card title="Add and configure agents" icon="sliders" href="/guides/add-and-configure-agents">
    The workflow for adding and tuning an agent.
  </Card>

  <Card title="Agent Marketplace" icon="store" href="/guides/agent-marketplace">
    Browse built-in, user-made, and community specializations.
  </Card>

  <Card title="Skills catalog" icon="list" href="/reference/skills-catalog">
    Lookup every built-in skill.
  </Card>

  <Card title="Permissions and safety" icon="shield" href="/reference/permissions-and-safety">
    Approval flows and safety controls.
  </Card>
</CardGroup>