---
title: "Zcode Review: The Official AI Coding Environment from Z.AI with GLM-5.2"
description: "Zcode is the official Agentic Development Environment from Z.AI for GLM-5.2. This review covers the new agent framework, Goal Mode, Bot Channels, remote control, and what it offers over Cursor and Claude Code."
date: 2026-07-03
categories: ["AI"]
tags: ["ai","coding"]
---

import { Image } from "astro:assets";
import Button from "@components/widgets/Button.astro";
import Notice from "@components/widgets/Notice.astro";
import ListCheck from "@components/widgets/ListCheck.astro";
import Accordion from "@components/widgets/Accordion.astro";
import Tabs from "@components/widgets/Tabs.astro";
import Tab from "@components/widgets/Tab.astro";

import zcodeInterface from "../../assets/images/26/02/zcode-interface.webp";
import zcodeModelsPanel from "../../assets/images/26/02/zcode-apikey-models-panel-en.webp";
import zcodePermissions from "../../assets/images/26/02/zcode-agent-permissions-en.webp";
import zcodeAgentsCall from "../../assets/images/26/02/zcode-agents-call-en.webp";
import zcodeSkillList from "../../assets/images/26/02/zcode-skill-list-en.webp";
import zcodeMcp from "../../assets/images/26/02/zcode-mcp.webp";
import zcodePluginDiscover from "../../assets/images/26/02/zcode-plugin-discover-en.webp";
import zcodeWelcome from "../../assets/images/26/07/zcode-welcome-workspace.webp";
import zcodeAgentFramework from "../../assets/images/26/07/zcode-agent-framework.png";
import zcodeExecutionModes from "../../assets/images/26/07/zcode-execution-modes.png";
import zcodeGoalMode from "../../assets/images/26/07/zcode-goal-mode.png";
import zcodeRemoteControl from "../../assets/images/26/07/zcode-remote-control.png";
import zcodeBotChannel from "../../assets/images/26/07/zcode-bot-channel.png";

Zcode is the official desktop development environment from [Z.AI](https://go.bitdoze.com/glm) — the company behind the GLM model family. It bundles a file manager, terminal, Git panel, live browser preview, and AI agent chat into one interface. Open a project, tell the agent what you need, and it reads your files, writes code, runs commands, and commits changes.

I've been testing it since it launched and the recent v3.2 update has made it a lot more stable. What makes it different from Cursor or Windsurf is the starting point. Those are editors with AI bolted on. Zcode starts from the agent and builds the interface around it. Z.AI calls it an **Agentic Development Environment**, which honestly sounds like marketing speak, but it's a fair description. The whole UI assumes you're talking to an agent, not editing files by hand.

<Notice type="info" title="Official Z.AI Tool">

Zcode is now the official development environment from Z.AI for GLM-5.2. It works with GLM Coding Plans and gives you **~1.5x the effective quota** when using Zcode compared to other tools. New users get a **5-day free trial** with 5M tokens/day (3M GLM-5.2 + 2M GLM-5-turbo).

</Notice>

## What Zcode does


<YouTubeEmbed
  url="https://www.youtube.com/embed/SJ6cA_fE5_Y"
  label="Zcode v3: The Official Z.AI Coding Environment with GLM-5.2"
/>


You interact through natural language, and the agent handles file operations, terminal commands, and browser previews. Here's the workspace you see when you open it:

<Image src={zcodeWelcome} alt="Zcode workspace showing task list, file tree, terminal, and agent chat" />

<ListCheck>

- **Zcode Agent**: The default agent built for GLM-5.2, with long-task planning, multi-turn context, and continuous code changes across files.
- **Goal Mode**: Set a verifiable session objective with `/goal` — the agent keeps iterating until the goal is verified complete.
- **Execution Modes**: Five modes from "ask before everything" to "full access" — cycle through them with Shift+Tab.
- **Multi-Model Support**: GLM-5.2, Claude, GPT, Kimi K2.5, DeepSeek, or any compatible model.
- **Bot Channel**: Connect WeChat or Feishu to control Zcode tasks from your phone's chat app.
- **Remote Control**: Scan a QR code to connect your phone to the desktop session and keep tasks moving.
- **AGENTS.md Memory**: Project-level instructions file that Zcode reads at the start of every task.

</ListCheck>

## Installation

Zcode is available as a desktop app for macOS, Windows, and Linux (beta).

### macOS

1. Download the DMG from [zcode.z.ai](https://zcode.z.ai/en/docs/install)
2. Open the DMG file
3. Drag **Zcode.app** to your **Applications** folder
4. Launch from Launchpad

### Windows

1. Download the installer from [zcode.z.ai](https://zcode.z.ai/en/docs/install)
2. Double-click the installer
3. Follow the setup wizard

### Linux (Beta)

Linux builds are available through the [Feishu beta group](https://applink.feishu.cn/client/chat/chatter/add_by_link?link_token=dabi8911-b0f9-41d8-8090-1b47c058a198). Both x64 and ARM64 AppImages are provided.

<Button
  text="Download Zcode"
  link="https://zcode.z.ai/en/docs/install"
  size="lg"
  color="blue"
  variant="solid"
/>

## Configuring AI providers

Zcode supports multiple providers. You're not locked into a single service.

### Supported providers

<Image src={zcodeModelsPanel} alt="Zcode models configuration panel with AI provider settings" />

| Provider | Models Available | How to Connect |
| --- | --- | --- |
| **Z.AI / BigModel** | GLM-5.2, GLM-5, GLM-4.7, GLM-4.6 | Built-in, bind your Zhipu account |
| **Anthropic** | Claude Sonnet 4.5, Opus 4.5, etc. | Enter Anthropic API key |
| **OpenRouter** | 100+ models | Set base URL to `https://openrouter.ai/api`, add API key |
| **Moonshot** | Kimi K2.5, Kimi K2-Turbo | Base URL: `https://api.moonshot.cn/anthropic` |
| **DeepSeek** | DeepSeek Chat | Base URL: `https://api.deepseek.com/anthropic` |
| **Custom** | Any OpenAI/Anthropic-compatible API | Add provider name, base URL, and API key |

<Tabs>
<Tab name="OpenRouter Setup">

1. Go to [openrouter.ai](https://openrouter.ai), create an account, and generate an API key
2. In Zcode, go to **Agents Settings > Models**
3. Click **Add Provider**
4. Set name to "OpenRouter"
5. Set API Base URL to `https://openrouter.ai/api`
6. Enter your API key
7. Enable the toggle

</Tab>
<Tab name="Anthropic Setup">

1. Get an API key from [console.anthropic.com](https://console.anthropic.com)
2. In Zcode, go to **Agents Settings > Models**
3. Select "Anthropic" from the provider list
4. Enable the toggle
5. Enter your API key

</Tab>
<Tab name="DeepSeek Setup">

1. Get an API key from DeepSeek
2. In Zcode, go to **Agents Settings > Models**
3. Click **Add Provider**
4. Set name to "deepseek"
5. Set Base URL to `https://api.deepseek.com/anthropic`
6. Enter your API key
7. Add model name "deepseek-chat"

</Tab>
</Tabs>

<Notice type="success" title="GLM Coding Plan Benefits">

If you use a [GLM Coding Plan](https://go.bitdoze.com/glm), Zcode consumes quota at a discounted rate — roughly **1.5x the effective allowance**. Off-peak hours (20 hours/day) use only 0.67x quota, and peak hours (14:00–18:00) use 2x instead of the usual 3x.

</Notice>

## Zcode Agent

The agent is the core of the experience. Zcode Agent is built in-house and deeply adapted for GLM-5.2. It handles complex project understanding, long-task planning, and continuous code changes across multiple files.

<Image src={zcodeAgentFramework} alt="Zcode Agent new task interface showing model picker, context, and execution modes" />

### Adding context

Click the **+** button at the bottom-left of the chat input to add context:

- **Attachments**: Upload screenshots, documents, and requirement material
- **@ mentions**: Reference files in the workspace so the agent can locate relevant code
- **# conversations**: Link a past conversation to bring its context into the current task
- **/ commands**: Call saved prompts to reuse a fixed workflow
- **$$ skills**: Invoke reusable skill playbooks

### Project instructions

Zcode reads an `AGENTS.md` file from your project root and loads it into every conversation. Put your project conventions, architecture notes, and coding standards here.

Two locations are supported:

| Source | Path | Scope |
| --- | --- | --- |
| User global | `~/.zcode/AGENTS.md` | All projects |
| Workspace | `AGENTS.md` in project root | Current project |

<Notice type="info" title="AGENTS.md replaces CLAUDE.md">

Zcode uses `AGENTS.md` instead of `CLAUDE.md`. If you have an existing Claude Code project, Zcode copies `CLAUDE.md` content into `AGENTS.md` during onboarding. After that, only `AGENTS.md` is read.

</Notice>

## Execution modes

Five modes control how much freedom the agent gets. Press **Shift + Tab** while the chat input is focused to cycle through them.

<Image src={zcodeExecutionModes} alt="Zcode execution modes selector showing Default, Confirm, Auto Edit, Plan, and Full Access" />

| Mode | What Happens | Good For |
| --- | --- | --- |
| **Default Mode** | Standard task strategy with balanced confirmations | Everyday development, routine edits |
| **Confirm Before Changes** | Asks before every file edit or command | Critical code, production configs |
| **Auto Edit** | Edits files automatically, commands still need confirmation | Routine iteration with fewer interruptions |
| **Plan Mode** | Creates a plan first, then implements after approval | Complex multi-step tasks |
| **Full Access** | Minimal interruptions, agent proceeds continuously | Low-risk tasks where you want speed |

## Goal Mode

This is one of the newer features and I've found it genuinely useful for long tasks. Type `/goal <objective>` and the agent keeps iterating toward the goal. After each round, it verifies whether the goal is met. If not, it continues automatically — no need to type "continue" every few minutes.

<Image src={zcodeGoalMode} alt="Goal Mode showing the agent iterating toward a verifiable objective with progress tracking" />

Good use cases for Goal Mode:

- "Refactor the whole module and keep the tests passing"
- "Fix all TypeScript compile errors"
- "Get this page's Lighthouse performance score above 90"

The summary panel in the top-right shows goal status, elapsed time, total tokens, and iteration count. You can pause, replace, or clear the goal at any time.

For long goal-driven tasks, pair Goal Mode with **Full Access** or **Auto Edit** to minimize interruptions and keep iterations flowing.

## Bot Channel

Bot Channel connects WeChat and Feishu to your Zcode workspace. After setup, you can message the bot from your phone's chat app to check task progress, send instructions, and keep the agent moving.

<Image src={zcodeBotChannel} alt="Bot Channel overview showing WeChat and Feishu integration options" />

Setting up Feishu:

1. Open **Bots** settings in Zcode's left sidebar
2. Click **Create Bot** and select Feishu
3. Scan the QR code — Zcode creates the Feishu app automatically
4. Send `/bind pairing-code` in the Feishu conversation
5. The bot is ready to use

WeChat works similarly — scan to sign in and bind automatically. DingTalk, Discord, and WeCom support are coming in later versions.

## Remote Control

Remote Control lets your phone connect to the current desktop workspace. Scan a QR code, and your phone shows a live view of the session. You can read agent responses, send new instructions, and confirm or stop long-running tasks.

<Image src={zcodeRemoteControl} alt="Remote Control entry point showing QR code dialog for mobile connection" />

The desktop stays as the runtime — the phone only handles display and text input. One phone connection per session. For longer-lived access from chat apps, use Bot Channel instead.

## Skills system

Skills are Markdown files that teach the agent how to handle specific tasks. They work like reusable playbooks.

<Image src={zcodeSkillList} alt="Zcode skills management panel showing installed user and plugin skills" />

| Type | Storage | Scope |
| --- | --- | --- |
| **User Skills** | `~/.claude/skills/` | All projects |
| **Plugin Skills** | Installed via plugins | Auto-available |
| **Project Skills** | `.claude/skills/` | Current project only |

## MCP services

Zcode uses the Model Context Protocol to extend what agents can do. Three built-in MCP services cover common needs:

<Image src={zcodeMcp} alt="Zcode MCP services configuration with visual understanding, web search, and web reader" />

<Tabs>
<Tab name="Visual Understanding">

**zai-mcp-server**: Lets the agent analyze images. Get an API token from [open.bigmodel.cn](https://open.bigmodel.cn), add your `Z_AI_API_KEY` in the MCP settings, and set `Z_AI_MODE` to `ZHIPU`.

</Tab>
<Tab name="Web Search">

**web-search-prime**: Gives the agent real-time internet access. Add your API key to the `Authorization` header as `Bearer your_api_key`.

</Tab>
<Tab name="Web Reader">

**web-reader**: Parses web pages and pulls out structured content. Same auth setup as web search.

</Tab>
</Tabs>

## Plugin system

Plugins add new commands, agents, MCP connections, language servers, and more. Zcode ships with access to the Claude Code plugin marketplace, so you can install community plugins you're already familiar with.

<Image src={zcodePluginDiscover} alt="Zcode plugin marketplace showing available plugins for agents, commands, MCP, and LSP" />

| Plugin Type | What It Adds |
| --- | --- |
| **Agent** | Specialized agents with domain workflows |
| **Command** | Custom `/` slash commands |
| **MCP** | External tool and data source connections |
| **LSP** | Language servers for code navigation and type checking |
| **Skill** | Behavior guidelines for the agent |
| **Hook** | Auto-execute actions on specific events |

## Version control

Zcode handles versioning at two levels.

**Conversation-level rollback**: Every message creates a checkpoint. Review all file changes after any interaction in a multi-file diff view, undo only the last interaction's changes, or jump back to the state after any specific message.

**Built-in Git panel**: View modified files with status markers, write commit messages and commit with one click, switch branches, and browse commit history.

## Zcode vs other AI coding tools

| Feature | Zcode | Cursor | Claude Code CLI | Windsurf |
| --- | --- | --- | --- | --- |
| **Type** | Standalone ADE | IDE (VS Code fork) | Terminal CLI | IDE (VS Code fork) |
| **Official Z.AI tool** | Yes | No | No | No |
| **Multi-Model** | GLM, Claude, GPT, Kimi, DeepSeek, custom | Multiple providers | Claude only | Multiple providers |
| **Goal Mode** | Yes, auto-iterate until verified | No | No | No |
| **Bot Channel** | WeChat, Feishu | No | No | No |
| **Skills System** | Yes, reusable Markdown playbooks | No | No | No |
| **Plugin Marketplace** | Agents, commands, MCP, LSP | Extensions | No | Extensions |
| **Chat Versioning** | Conversation checkpoints + rollback | No | No | No |
| **Remote Dev (Mobile)** | QR code + Bot Channel | No | No | No |
| **AGENTS.md** | Auto-loaded per project | .cursorrules | CLAUDE.md | .windsurfrules |

## Getting started in 5 minutes

**Step 1**: Download and install Zcode from [zcode.z.ai](https://zcode.z.ai/en/docs/install)

**Step 2**: Open the app and click "Connect" to set up your AI provider. If you have a [GLM Coding Plan](https://go.bitdoze.com/glm), bind it for the best experience with GLM-5.2 and the 1.5x quota benefit.

**Step 3**: Open a project folder in Zcode.

**Step 4**: Type `/init` or create an `AGENTS.md` file in your project root with your conventions.

**Step 5**: Start chatting. Ask the agent to explain your codebase, fix a bug, or build a new feature.

<Button
  text="Get Started with Zcode"
  link="https://zcode.z.ai/en/docs/install"
  size="lg"
  color="blue"
  variant="solid"
/>

## GLM Coding Plan pricing

Z.AI offers GLM Coding Plans that work with Zcode. The plans give you access to GLM-5.2 and other GLM models with generous token allowances.

<Notice type="success" title="Exclusive Benefits for Zcode Users">

- **5-day free trial** for new Zcode users: 3M tokens/day of GLM-5.2 + 2M tokens/day of GLM-5-turbo (5M total)
- **~1.5x effective quota** for GLM Coding Plan subscribers using Zcode — off-peak hours consume only 0.67x quota
- These benefits run through **July 31, 2026**

</Notice>

<Button
  text="Get GLM Coding Plan (10% OFF)"
  link="https://go.bitdoze.com/glm"
  size="lg"
  color="blue"
  variant="solid"
/>

## Pros and cons

**What I like:**

- GLM-5.2 integration is deep — the agent, model, and execution workflow are tuned together
- Goal Mode saves me from typing "continue" every 5 minutes on long tasks
- Bot Channel means I can check on tasks from WeChat without opening my laptop
- The 1.5x quota benefit with GLM Coding Plans makes it cheaper to run than alternatives
- Linux support (beta) — wasn't there at launch

**What I don't:**

- The plugin ecosystem is still small compared to Cursor's extension marketplace
- Some MCP services need Zhipu/BigModel API keys specifically
- The learning curve is real — agents, skills, plugins, MCP, Goal Mode, execution modes. There's a lot to figure out.
- Bot Channel only supports WeChat and Feishu right now (Discord and Slack coming later)

## Frequently asked questions

<Accordion label="Is Zcode free?" group="faq" expanded="true">

Zcode itself is free to download and use. You pay for the AI models you connect — either through Z.AI's GLM Coding Plans, OpenRouter credits, Anthropic API usage, or whichever provider you choose. New users get a 5-day free trial with 5M tokens/day.

</Accordion>

<Accordion label="Do I need Claude Code to use Zcode?" group="faq">

No. Zcode supports multiple AI providers. You can use GLM-5.2, models through OpenRouter, the Anthropic API directly, Moonshot (Kimi), DeepSeek, or any service that speaks the OpenAI or Anthropic protocol.

</Accordion>

<Accordion label="What is the GLM Coding Plan quota benefit?" group="faq">

GLM Coding Plan subscribers using Zcode get roughly 1.5x the effective token allowance. Off-peak hours (20 hours/day) consume only 0.67x quota per token. Peak hours (14:00–18:00) consume 2x instead of the usual 3x. This means the same plan lasts longer in Zcode than in other tools.

</Accordion>

<Accordion label="Does Zcode work on Linux?" group="faq">

Yes, Linux is now supported in beta. Both x64 and ARM64 AppImages are available. You can join the Linux beta group through the Feishu link on the download page.

</Accordion>

<Accordion label="Can I use Zcode with local models?" group="faq">

Yes. Zcode supports any OpenAI or Anthropic-compatible API endpoint. Point it at a local inference server (like Ollama or vLLM) and it'll work with whatever model you're running.

</Accordion>

<Accordion label="How does Zcode compare to Cursor?" group="faq">

Cursor is a VS Code fork with AI editing features. Zcode is a standalone environment built around agent interactions. Zcode has Goal Mode, Bot Channel, specialized agents, a skills system, and conversation-level version rollback. Cursor has a more familiar IDE experience and a larger community.

</Accordion>

## Wrapping up

Zcode has come a long way since its initial launch. The v3.2 update added Linux support, Goal Mode, Bot Channels for WeChat and Feishu, and a bunch of stability fixes. As Z.AI's official coding environment for GLM-5.2, it gets the deepest integration with their models and the quota benefits that come with it.

The Goal Mode feature alone has changed how I handle long tasks — set it and walk away. Come back to a completed result. The Bot Channel is useful for checking on progress from your phone without needing to be at your desk.

If you're already using GLM models or want to try them, Zcode is the way to go. The 5-day free trial and 1.5x quota benefit make it worth testing.

<Button
  text="Try Zcode Free"
  link="https://zcode.z.ai/en/docs/install"
  size="lg"
  color="blue"
  variant="solid"
/>