Qwen Code Developer Guide
Qwen Code is an open-source agentic coding assistant built for terminal-first software workflows. This guide explains how to think about Qwen Code as a developer tool, how its daemon and permission system work, and where to start in the repo.
This guide is based on the official Qwen Code docs and the official QwenLM/qwen-code repository, checked on June 26, 2026.
In this section​
- Setup and first run
- Architecture and runtime
- Configuration and security
- Extending and contributing
- Simple user guide
1. What Qwen Code is​
The shortest accurate description is:
Qwen Code is a terminal-based coding agent that combines LLM reasoning, repo-aware actions, and a structured permission system for software tasks.
That places it directly in the same category as tools like Codex CLI, Gemini CLI, Claude Code, and Crush.
2. The mental model​
Think of Qwen Code as five cooperating pieces:
| Layer | What it does |
|---|---|
| terminal UI | Main user interaction |
| daemon/runtime | Long-lived execution and coordination |
| permission mediation | Controls risky actions |
| model and context layer | Provides reasoning and repo context |
| extension and workflow surfaces | Support broader integration and use cases |
3. Why developers would choose it​
- It is explicitly built for agentic coding.
- The docs expose internal topics like daemon architecture and permission mediation, which makes it easier to reason about than a black-box assistant.
- It is open source and terminal-first.
- It supports a workflow that is serious enough for real coding, not only small chat prompts.
4. What makes the docs useful​
Qwen Code's docs are notable because they do not stop at quickstart material. They also explain developer-facing internals like daemon architecture and permission mediation, which are exactly the right topics for a trustworthy coding agent.
5. Recommended reading order​
- this overview,
- quickstart,
- commands and settings,
- daemon and permission docs,
- repo structure once the user flow is familiar.