A development workflow system with more than 15 commands and specialized AI roles for handling software tasks. The available description identifies it as an intelligent development-automation tool, but does not explain each command or role.
38 5mo agoA
tokens not measured
originalApache-2.0
A helper that independently checks whether a reported software bug was fixed correctly. It reviews the fix, code quality, regression risks, and testing, with simpler checks for front-end changes.
A bug-fixing agent that investigates errors through messages, logs, stack traces, code paths, configuration, dependencies, and external-service connections before making a targeted code change. It also recognizes likely front-end issues in files and frameworks such as React, Vue, and Svelte.
A development coordinator that guides work from a software requirement to working code through design, implementation, integration, review, and testing. It also considers error handling, compatibility, maintainability, and future scaling.
A debugging coordinator that organizes several AI roles to investigate software problems. It combines architecture analysis, external research, coding, testing, and a final diagnosis before asking for confirmation to proceed with a fix.
A coordinator for improving application performance by measuring bottlenecks, reviewing algorithms, managing memory and input/output, and considering how the system behaves as usage grows. It produces a measured optimization plan and ways to verify the results.
A command-driven workflow for diagnosing and fixing software bugs, especially problems in frontend code such as React, Vue, or Angular. It analyzes the cause, applies a fix, and runs a separate verification step that may send the work back for revision.
A command that examines Git changes and suggests a commit message, with optional task or bug references and Chinese-language output. Git is a version-control system that records changes to a codebase; the command shows a preview and asks for confirmation before committing and pushing.
A command that reads the changes already placed in Git’s staging area—the files selected with git add—and suggests a commit message. It can follow common message formats, choose English or Japanese, and flag breaking changes.
A command for investigating and planning software performance improvements. It starts by measuring bottlenecks, then considers algorithms, memory, input/output, system resources, and how the software behaves as usage grows.
A command that builds or updates AI guidance documents named CLAUDE.md at the repository root and inside detected modules. The documents describe the project structure, important files, interfaces, dependencies, entry points, tests, and project rules.
A command for carrying out planned code restructuring, such as splitting large methods or classes, renaming code, and correcting violations of SOLID design principles. It is meant for gradual changes that keep the software working the same way.
A command for reviewing a selected part of a codebase from several angles, including code quality, security, speed, and overall structure. It produces findings, suggested improvements, and an action plan.
A macOS-only command that captures a window, the full screen, or a selected area and then makes the image available for analysis. It uses macOS’s built-in screencapture tool and does not itself take a screenshot without being asked.
A command for searching the web through the Gemini command-line tool, which is a terminal program for interacting with Google’s Gemini service. It is intended for finding current information, researching documentation, investigating errors, and comparing technologies.
Use for any development task. Always starts with brainstorming to clarify requirements. Frontend tasks then go through design-first workflow (Visual Companion sketch → ui-ux-pro-max design system → writing-plans → execute → verification). Non-frontend tasks go directly to superpowers standard workflow (writing-plans →…