Human-in-the-Loop Task Manager for AI Agents
Checked October 30, 2026 | All YOLO Mode guides

How to Enable YOLO Mode in Augment Code

Auggie, Augment Code's command-line agent, has no YOLO flag and no auto-approve mode in its docs. What decides whether a tool call runs, asks or is blocked is a list of tool permission rules, set in settings.json or with --permission. Allow everything and you are close to YOLO. Mark tools as ask-user and, connected to AgentRQ, you get a YOLO switch you can flip per task.

This page is one of our YOLO mode guides, and part of our guide to YOLO mode in every coding agent. Everything below was checked against Augment's Auggie docs (docs.augmentcode.com, 0.x and the 1.x preview) and the Auggie CLI package (@augmentcode/auggie 0.36.0, the version the ACP registry ships) 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

Does Augment Code Have a YOLO Mode?

Not as a switch. Each rule names a tool and gives it one of five permission types:

Type Effect
allow Runs without asking
ask-user Asks you first. With no interactive UI, the call is denied
deny Blocked
webhook-policy An HTTP endpoint decides
script-policy A local script decides

Within one file the first matching rule wins. Across files and --permission, the most restrictive answer wins, so an allow in your own settings can never override a deny from a project or managed file. When no rule matches, the docs only say the CLI "follows its implicit runtime behavior".

Rules are read from ~/.augment/settings.json, .augment/settings.json in the repo, .augment/settings.local.json and a managed file in /etc/augment/. /permissions in an interactive session shows the active rules and lets you edit them. The rules apply to the Auggie CLI and Augment's Cosmos cloud agents, not to the Augment editor extension. Auggie's docs describe no local sandbox for the CLI.

The Closest Thing to YOLO in Auggie

Allow the tools you want to run unattended. Auggie 0.36.0 uses these tool names:

json 
{
  "toolPermissions": [
    { "toolName": "view", "permission": { "type": "allow" } },
    { "toolName": "str-replace-editor", "permission": { "type": "allow" } },
    { "toolName": "save-file", "permission": { "type": "allow" } },
    { "toolName": "launch-process", "permission": { "type": "allow" } }
  ]
}

The Auggie 1.x preview renames them to read, edit, write and bash, and still accepts the old names. Note that permission must be an object with a type. A bare string such as "allow" is dropped.

The Halfway Option

Shell rules can match the command with shellInputRegex. Allow the commands you trust and ask about the rest:

json 
{
  "toolPermissions": [
    { "toolName": "launch-process", "shellInputRegex": "^(npm test|npm run lint|git status|git diff)", "permission": { "type": "allow" } },
    { "toolName": "launch-process", "permission": { "type": "ask-user" } }
  ]
}

For a single run, add rules on the command line, for example auggie --permission launch-process:ask-user.

How to Enable YOLO Mode in Augment Code Through AgentRQ

Auggie's rules are fixed for a whole run. AgentRQ lets you decide per task, and answer from somewhere other than the terminal.

Approve From the AgentRQ Task Instead of the Terminal

Connect Auggie to an AgentRQ workspace through the ACP Gateway (Augment Code setup guide):

bash 
npx @agentrq/acp-gateway@latest --agent auggie

In ACP mode, Auggie sends a permission request only when an ask-user rule matches the tool call. So mark the tools you want to review as ask-user, for example editing, writing and shell commands, and leave reading on allow. Each of those requests now shows up in the AgentRQ task on the web, on your phone or in Slack, together with the command or edit it is asking about. Answer Allow Once, Always Allow or Deny from wherever you are, and Auggie carries on. If nobody answers within 30 minutes, the gateway cancels the turn instead of guessing. Change the limit with --permission-timeout.

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. Every permission request in it is approved automatically and Auggie never waits, while every other task in the workspace still asks. This is the YOLO switch Auggie itself does not have. Scheduled tasks, event triggers and workflow 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

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

A Safe YOLO Setup for Augment Code

  • → Keep deny rules for the dangerous commands. A deny wins over everything, including AgentRQ's YOLO, because Auggie never asks about a denied call.
  • → 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. The Auggie CLI has none, 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 lists every call the task made, including the ones that were auto-approved.

FAQ

Does Augment Code have a YOLO mode? No. Auggie has no YOLO flag. Tool permission rules decide which tools run, ask or are blocked.

How do I let Auggie run tools without asking? Add allow rules for those tools under toolPermissions in ~/.augment/settings.json, or pass --permission rules for one run.

How do I make Auggie ask before running shell commands? Add a rule that gives the shell tool the ask-user type. In Auggie 0.36.0 that tool is launch-process, in the 1.x preview it is bash.

Does Augment Code have a sandbox? The Auggie CLI docs describe none. Run unattended sessions in a container or a throwaway clone.

How do I approve Auggie commands remotely? Connect Auggie to AgentRQ through the ACP Gateway and mark the tools you want to review as ask-user. 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.

Start Free