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

# Harness overview

> Understand the Voss terminal harness and how it differs from open-ended AI chat.

The Voss harness is a terminal-native agent loop for bounded, inspectable, resumable repo work. It is the **single-agent substrate** the [orchestration layer](/orchestration/overview) builds on: a team run is many of these loops, each a scoped node in a [session tree](/orchestration/session-tree).

It is not a general chat box. Each turn goes through a small set of explicit systems:

```text theme={"theme":"github-dark"}
user task
  -> provider plan
  -> confidence gate
  -> permission gate
  -> tool execution
  -> session record
```

## Why it exists

AI coding tools are useful, but many hide important decisions in prompts:

* What can the agent edit?
* Which commands can it run?
* What context did it inspect?
* Which files did it avoid?
* Can the session be resumed later?

Voss makes those decisions explicit through commands, modes, tool classification, permission prompts, persistent project context, and resumable sessions.

## Main commands

<CardGroup cols={2}>
  <Card title="voss do" icon="terminal" href="/harness/commands">
    Run one natural-language task and exit.
  </Card>

  <Card title="voss chat" icon="messages-square" href="/harness/commands">
    Start an interactive harness REPL.
  </Card>

  <Card title="voss edit" icon="file-pen-line" href="/harness/scoped-editing">
    Open a scoped edit session around explicit targets.
  </Card>

  <Card title="voss sessions" icon="history" href="/harness/sessions">
    List and resume prior harness work.
  </Card>

  <Card title="voss team run" icon="workflow" href="/orchestration/overview">
    Run a goal as a whole team built from these loops.
  </Card>
</CardGroup>

## Product shape

The harness is the substrate; the [orchestration layer](/orchestration/overview) is the product built on top of it. One harness loop is a single bounded agent. A team run dispatches many of them under an [Engineering Manager loop](/orchestration/em-loop), each owning a scoped, budgeted node in the [session tree](/orchestration/session-tree), gated by independent [review](/orchestration/reviewers).

The Python harness is the canonical behavior source and the npm CLI vendors that implementation. The `.voss` language is the control layer at both levels — confidence, context budgets, routing, memory, tools, and fallbacks for one agent; [roles, principles, and gates](/language/core-constructs) for a team.