---
title: "How to Enable YOLO Mode in Dirac | AgentRQ Guidelines"
url: https://agentrq.com/guidelines/yolo-mode/dirac
description: "Dirac YOLO mode: dirac --yolo (-y) for unattended plain-output runs, or --auto-approve-all to keep the interactive UI. Defaults, tool limits and AgentRQ."
---

![](https://agentrq.com/assets/agent-icons/dirac.svg) Checked October 30, 2026 | [All YOLO Mode guides](https://agentrq.com/guidelines/yolo-mode/)

# How to Enable YOLO Mode in Dirac

To run Dirac in YOLO mode, start it with `dirac --yolo` (or `-y`). That approves every action and also switches Dirac to plain, non-interactive output, which suits scripts and CI. If you want auto-approval but still want the interactive terminal UI, use `dirac --auto-approve-all` instead.

This page is one of our [YOLO mode guides](https://agentrq.com/guidelines/yolo-mode/), and part of our guide to [YOLO mode in every coding agent](https://agentrq.com/blog/what-is-yolo-mode-and-how-to-turn-it-on-in-every-coding-agent). Everything below was checked against the Dirac source and CLI README (dirac-run/dirac, version 0.5.24) on October 30, 2026.

![Animated sketch of YOLO mode in AgentRQ: claude waits to run go test, YOLO is flipped on for the task, and every later edit and test run lands auto-allowed in the Trajectory view](https://agentrq.com/assets/sketches/yolo-mode.svg)

## How to Enable YOLO Mode in Dirac

| Flag | Approvals | Interface |
| --- | --- | --- |
| `dirac --yolo` (or `-y`) | all approved | plain output, exits when done |
| `dirac --auto-approve-all` | all approved | interactive terminal UI |
| no flag | asks | interactive UI, or an error in plain mode |

bash

```bash
dirac --yolo "Run the tests and fix failures"
```

`--yolo` is also how Dirac gets through a run with nobody watching. Without it, a plain-output run (piped input, redirected output or `--json`) exits with an error the first time it needs an approval. `--max-consecutive-mistakes <count>` sets how many mistakes in a row a YOLO run tolerates before it stops.

Neither flag is saved. They apply to that run only.

### Limit What a YOLO Run Can Touch

Dirac lets you choose the tools for one run, which pairs well with `--yolo`:

bash

```bash
dirac --yolo --enable-tool read_file,edit_file --disable-tool use_subagents "Fix the issue"
dirac --only-tools read_file,search_files,respond "Analyze the workspace"
```

`dirac tools` prints your current tool settings as ready-to-paste flags. These options do not work with `--acp`.

### The Halfway Option

Dirac is a Cline descendant and keeps per-action auto-approve settings. Out of the box it already approves reading files in the workspace and commands it judges harmless, and asks before editing files, using the browser, or touching files outside the workspace. You can widen or narrow this under `/settings`.

Dirac can also hand approvals to a second model first. Set a **Utility model** and give it a plain-language permission policy: it approves requests that meet the policy and passes the unsafe or unclear ones to you, and every automatic approval shows in the transcript with its reason.

## What YOLO Mode Turns Off in Dirac

Approval prompts, and in the case of `--yolo`, the interactive UI. Dirac's docs describe no sandbox, so there is nothing else to drop. Plan mode (`--plan`) is separate: it gathers information and writes a plan instead of acting.

## How to Enable YOLO Mode in Dirac Through AgentRQ

`--yolo` covers a whole run. AgentRQ lets you make the decision per task, and answer from somewhere other than the terminal.

### Approve From the AgentRQ Task Instead of the Terminal

Connect Dirac to an AgentRQ workspace through the [ACP Gateway](https://agentrq.com/docs/connect-acp-agent) ([Dirac setup guide](https://agentrq.com/docs/agents/dirac)):

bash

```bash
npx @agentrq/acp-gateway@latest --agent dirac
```

The gateway starts `dirac --acp` without `--yolo` or `--auto-approve-all`, so YOLO is off. Each approval Dirac would have asked you for in the terminal shows up in the AgentRQ task on the web, on your phone or in Slack, together with the command or edit it is about. Answer **Allow Once**, **Always Allow** or **Deny** from wherever you are, and Dirac carries on. If nobody answers within 30 minutes, the gateway cancels the turn instead of guessing. Change the limit with `--permission-timeout`.

Dirac's two ACP session modes are **Plan** and **Act**, not approval levels, so AgentRQ leaves the mode alone. The actions Dirac auto-approves by default, such as reading workspace files and harmless commands, still run without asking.

### Turn On YOLO for One Task

Flip the **YOLO** toggle on a task, or when you create it, to put just that task in [YOLO mode](https://agentrq.com/features/yolo-mode). Every permission request in it is approved automatically and Dirac never waits, while every other task in the workspace still asks. [Scheduled tasks](https://agentrq.com/features/task-scheduling), [event triggers](https://agentrq.com/features/events) and [workflow](https://agentrq.com/features/workflows) steps have the same switch, so an unattended job can run in YOLO while your interactive work stays supervised.

![The YOLO toggle on a task in AgentRQ being switched on, then off again](https://agentrq.com/assets/guidelines/yolo-mode-task-toggle.gif)

### Turn YOLO On or Off in Workspace Settings

To make YOLO the default for a whole workspace, open the workspace, go to **Settings**, choose the **Automations** tab and flip **YOLO Mode (Execute All)**. Click **Update Workspace** to save. Switch it off the same way. The switch is off by default, and every permission request in the workspace is approved automatically while it is on.

![Workspace Settings in AgentRQ: the YOLO Mode (Execute All) switch turned on and saved, then off and saved](https://agentrq.com/assets/guidelines/yolo-mode-workspace-settings.gif)

### A Safe YOLO Setup for Dirac

- → **Allow the boring tools, not everything.** **Always Allow** adds the tool to the workspace's auto-approved list, so the routine calls stop asking and the risky ones still do. Prune that list in workspace settings now and then.
- → **Leave the workspace-wide switch off.** Turn YOLO on per task instead, so it never outlives the job you trusted.
- → **Bring your own sandbox.** Dirac does not isolate the commands it runs, so run YOLO tasks in a container or a throwaway clone.
- → **Commit first.** Start every YOLO task from a clean git tree, so any change is one `git reset` away.
- → **Read the record.** The [tool call history](https://agentrq.com/features/tool-call-history) lists every call the task made, including the ones that were auto-approved.

## FAQ

**How do I enable YOLO mode in Dirac?** Run dirac --yolo, or -y. It approves every action and uses plain, non-interactive output.

**What is the difference between --yolo and --auto-approve-all in Dirac?** Both approve every action. --yolo switches to plain output for unattended runs, while --auto-approve-all keeps the interactive terminal UI.

**Why does Dirac exit with an error in a script?** Without --yolo, a plain-output run exits with an error when it needs an approval or your feedback.

**Can a Dirac YOLO run use only some tools?** Yes. Combine --yolo with --enable-tool, --disable-tool or --only-tools for that run.

**How do I approve Dirac actions remotely?** Connect Dirac to [AgentRQ](https://agentrq.com) through the ACP Gateway. Its permission requests show up in the task on the web, your phone or Slack.

For every other agent, see [what YOLO mode is and how to turn it on in every coding agent](https://agentrq.com/blog/what-is-yolo-mode-and-how-to-turn-it-on-in-every-coding-agent).