zhukunpenglinyutong/desktop-cc-gui
Skill Claude CodeCodex
A plugin-building workflow for CC GUI, a desktop client that can be extended with plugins written as configuration or JavaScript modules.
Agent Claude Code
A software architecture guide for planning how an application is organised, changed, and scaled. It reviews existing systems and records design decisions, trade-offs, and possible future problems.
Agent Claude Code needs its repo
A coding agent for fixing TypeScript, compilation, dependency, and build-configuration errors. TypeScript is a programming language that checks types before code runs, while a build turns source code into a runnable application.
Agent Claude Code
An automated code-review role that checks changed code for readability, security, maintainability, performance, testing, and library-license concerns. Its review output groups findings by severity and includes file locations and suggested fixes.
Agent Claude Code
An AI agent that keeps a project's documentation and code maps aligned with its source code. A code map is a guide to the project's folders, modules, dependencies, routes, and data flow.
Agent Claude Code needs its repo
An end-to-end testing agent built around Playwright, a tool that drives real browsers through user journeys. It creates, maintains, runs, and reports on these tests.
Agent Claude Code
A planning agent for complex features, architecture changes, and major refactoring. It turns a broad coding request into a detailed implementation plan.
Agent Claude Code
A code-maintenance workflow that finds unused code, duplicate code, unused dependencies, and exports that are no longer needed, then removes them carefully.
Agent Claude Code
A security-review guide for finding and fixing common web application vulnerabilities, including exposed secrets, unsafe input handling, broken access control, and risky dependencies.
Agent Claude Code
A test-driven development guide, where tests are written before the code they check.
Command Claude Code
An automatic command selector that reads a described task and routes it to a matching development command.
Command Claude Code
A command that incrementally fixes TypeScript and build errors by running the project build, grouping and explaining errors, applying fixes, and checking the result again.
Command Claude Code
A review command that examines uncommitted code changes for security problems, code-quality issues, and common development mistakes.
Command Claude Code
A command that uses Playwright to create and run end-to-end tests, which check complete user journeys through an application.
Command Claude Code
A planning command that restates a development request, identifies risks, and creates a step-by-step implementation plan. It waits for the user's approval before code changes begin.
Command Claude Code
A dead-code cleanup command for finding code that is no longer used and preparing safe removals. Dead code means files, exports, or dependencies that the project does not use.
Command Claude Code
A command that uses test-driven development (TDD): define the code interface, write tests first, implement the smallest working solution, and then improve it. It also checks that at least 80% of the code is covered by tests.
Command Claude Code
A command that checks automated-test coverage and creates tests for code paths that are not tested. Test coverage measures which parts of a program are exercised by tests.
Command Claude Code
A command that scans a codebase and updates concise architecture documents describing its backend, frontend, data model, and overall structure.
Command Claude Code
A command that creates developer documentation from project settings and environment examples. It uses package.json, a file that defines project scripts, and .env.example, a template listing configuration variables.
Hook Claude Code
One of 3 in hooks.json
Runs before the agent uses a tool for tool == "Bash" && tool_input.command matches "(npm run dev, pnpm( run)? dev, yarn dev, bun run dev)", tool == "Bash" && tool_input.command matches "(npm (install, test), pnpm (install, test), yarn (install, test), bun (install, test), cargo build, make, docker, pytest, vitest…
Hook Claude Code
One of 3 in hooks.json
Runs after a tool call finishes for tool == "Bash", tool == "Edit" && tool_input.file_path matches "\\.(ts, tsx, js, jsx)$", tool == "Edit" && tool_input.file_path matches "\\.(ts, tsx)$", tool == "Edit" && tool_input.file_path matches "\\.(ts, tsx, js and jsx)$" tool calls, running #!/bin/bash (4 commands). From…
Hook Claude Code
One of 3 in hooks.json
Runs when the agent finishes a response, running #!/bin/bash. From zhukunpenglinyutong/ai-max.
Skill Claude CodeCodex
A security checklist for code that handles login permissions, user input, uploaded files, passwords, API keys, payments, or other sensitive data.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: