How to Enable YOLO Mode in GLM Agent

GLM Agent has no command line and no --yolo flag. It only runs behind an ACP client, and its YOLO mode is a session mode the client picks: Bypass all permissions (bypass_permissions). In that mode file writes, shell commands and web reads all run without asking.

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 the GLM Agent source and README (stefandevo/glm-acp-agent, version 1.17.0) 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

GLM Agent's Three Permission Modes

GLM Agent advertises three ACP session modes. Every new session starts in the first one.

Mode ID	Name	File writes and edits	Shell commands	Web reader, remote images
`default`	Ask for permission	ask	ask	ask
`accept_edits`	Auto-approve edits	run	ask	ask
`bypass_permissions`	Bypass all permissions	run	run	run

Reading files, listing directories, web search and the todo list never ask, in any mode. MCP tools passed in by the client don't ask either.

How to Enable YOLO Mode in GLM Agent

Pick Bypass all permissions in your editor's mode selector. Under the hood the client calls session/set_mode with bypass_permissions, and GLM Agent stops sending permission requests for that session. There is no environment variable or config file for it: of the ACP_GLM_* variables, none sets the mode. A session saved to disk keeps its mode when it is loaded again.

The Halfway Option

Auto-approve edits (accept_edits) lets GLM Agent write and edit files freely but still asks before every shell command and every web fetch. When it does ask, it offers only Allow and Skip, so there is no "always allow" memory of its own.

What YOLO Mode Turns Off in GLM Agent

Only the prompts. GLM Agent has no sandbox to turn off. Its README says commands run with the agent process's own user permissions and network access, and that permission prompts govern consent, not isolation. File reads and directory listings are confined to the session's workspace, but shell commands are not.

How to Enable YOLO Mode in GLM Agent Through AgentRQ

Because GLM Agent's YOLO is a session mode, it lives in whichever client happens to be running it. AgentRQ lets you make that call per task instead, and answer prompts from anywhere.

Approve From the AgentRQ Task Instead of the Terminal

Connect GLM Agent to an AgentRQ workspace through the ACP Gateway (GLM Agent setup guide):

bash

npx @agentrq/acp-gateway@latest --agent glm-acp-agent

GLM Agent starts every session in default, Ask for permission, and AgentRQ keeps it there: it picks the mode that asks and switches the session back if it leaves. So every file write, shell command and web fetch shows up in the AgentRQ task on the web, on your phone or in Slack, with the full arguments GLM Agent wants to run. Answer Allow Once, Always Allow or Deny from wherever you are, and GLM Agent 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 GLM Agent never waits, while every other task in the workspace still asks. 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 GLM Agent

→ 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. GLM Agent asks about only four tools (write_file, edit_file, run_command and web_reader, plus remote images), so the list stays short. Prune it 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. GLM Agent runs shell commands as your user, with your network. 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

How do I enable YOLO mode in GLM Agent? Switch the session to the Bypass all permissions mode (bypass_permissions) in your ACP client. GLM Agent has no command-line flag for it.

What does Auto-approve edits do in GLM Agent? It lets file writes and edits run without asking, while shell commands and web fetches still ask.

Which GLM Agent tools never ask for permission? Reading files, listing directories, web search, the todo list and MCP tools from the client run without a prompt in every mode.

Does GLM Agent have a sandbox? No. Its shell commands run with your user's permissions and network. Run YOLO sessions in a container or a throwaway clone.

How do I approve GLM Agent commands remotely? Connect GLM Agent to AgentRQ 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.