e2e-testing

Test GLM Code CLI end-to-end with real model calls and API traffic inspection.

Updated May 15, 2026
One-click install
npx skills add https://github.com/qazz92/glm-code --skill e2e-testing-qazz92
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: e2e-testing
Source: https://github.com/qazz92/glm-code/tree/main/.glm/skills/e2e-testing
Command: npx skills add https://github.com/qazz92/glm-code --skill e2e-testing-qazz92

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill unit provides comprehensive guidance for running end-to-end tests of the GLM Code CLI, ensuring that the CLI behaves as expected with real model calls.

Core Features & Use Cases

  • End-to-End Testing: Facilitates running tests from building the CLI bundle to inspecting raw API traffic.
  • Headless Mode: Supports non-interactive execution with JSON output, ideal for automated testing.
  • API Traffic Inspection: Enables debugging model behavior by enabling API logging to inspect request/response payloads.
  • Interactive Mode (tmux): Allows verifying TUI rendering and testing keyboard interactions.
  • MCP Server Testing: Provides instructions for setting up and running end-to-end tests involving MCP tool servers.
  • Token Usage Stats: Summarizes token usage across recent API logs to verify prompt caching and investigate unexpected token counts.

Quick Start

Run the e2e-testing skill to verify the behavior of the GLM Code CLI with real model calls.

Frequently Asked Questions about e2e-testing

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run end-to-end tests for GLM Code CLI with real model calls?

Run end-to-end tests for GLM Code CLI by building the CLI bundle, executing commands in headless mode for JSON output, and inspecting raw API traffic. This skill provides step-by-step guidance to verify CLI behavior with real model calls.

What is headless mode and how does it help with CLI testing?

Headless mode supports non-interactive execution with JSON output, making it ideal for automated end-to-end testing. It allows you to verify GLM Code CLI behavior without manual input, enabling seamless integration into automated testing pipelines.

How can I inspect API traffic to debug model behavior in GLM Code CLI?

Inspect API traffic by enabling API logging within the GLM Code CLI. This allows you to examine raw request and response payloads, debug model behavior, and reproduce user-reported bugs by analyzing the exact data exchanged with the model.

How do I test MCP server integrations with the GLM Code CLI?

Test MCP server integrations by following the skill's instructions for setting up and running end-to-end tests involving MCP tool servers. This ensures your MCP tool integrations function correctly and interact properly with the CLI environment.

Can I verify TUI rendering and keyboard interactions during testing?

Yes, you can verify TUI rendering and test keyboard interactions using interactive mode with tmux. This allows you to simulate user input and visually confirm that the terminal user interface renders correctly during end-to-end testing.

How do I check token usage and prompt caching behavior in API logs?

Check token usage by summarizing token counts across recent API logs. This feature helps you verify prompt caching effectiveness and investigate unexpected token consumption during end-to-end testing with real model calls.