Skip to content
Get packs

End-to-End Testing for Qwen Code

Run and inspect real end-to-end tests of the Qwen Code CLI, including MCP and API traffic

Source Download skill
Repository
QwenLM/qwen-code
Stars
28,880
Licence
Apache-2.0
Files
6 files

What it does

This skill gives an AI coding agent a playbook for testing the Qwen Code command-line tool the way a real user would: running it against a live model, driving its interface in a terminal session, and recording the raw requests it sends to the model. It also includes small helper scripts for building a fake MCP tool server, a fake OpenAI-style model server for scripted scenarios, and a summary of token usage from logged requests. Without the skill, an agent would have to work out these steps from scratch.

Great for

  • Maintainers of Qwen Code reproducing a user-reported bug end to end
  • Contributors verifying that a fix works in the built bundle with real model calls
  • Developers testing how the CLI integrates with a custom MCP tool server

Skip it if

  • Ordinary users of Qwen Code or other projects, since it only makes sense inside the Qwen Code repository and its setup
  • Projects that do not use Qwen Code, because the commands and paths are specific to it
  • Running against real accounts on shared machines without care, because the tests use real API keys from ~/.qwen and can spend money or write chat history

Try asking

  • “Reproduce this bug report end to end using the globally installed qwen command and show me the tool errors”
  • “Test my MCP echo server integration with the CLI in headless mode and check the tool_use output”
  • “Capture the raw API request and response for this prompt so I can see the tool schema sent to the model”
  • “Summarize token usage and cache hit rate for the last 20 requests”
Step-by-step guide: how to use End-to-End Testing for Qwen CodeInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 6 original files, the Apache-2.0 licence and an attribution note.

    e2e-testing.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip e2e-testing.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip e2e-testing.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Use QWEN_RUNTIME_DIR to keep test runs from cluttering your real chat history and auto-memory, unless the bug depends on that history
  • Point the CLI at the included mock server via OPENAI_BASE_URL for deterministic tests that do not spend real model tokens
  • Read the scripts before running them, since they execute on your machine and the mock server listens on a local port

What's inside

  • references/mcp-testing.md
  • references/mock-openai-server.md
  • scripts/mcp-test-server.js
  • scripts/mock-openai-server.js
  • scripts/token-stats.py
  • SKILL.md

Redistributed unmodified from QwenLM/qwen-code under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.