auto-work
01Command Claude Code
A command that takes a requirement through research, planning, plan review, development, and development review.
Automatically completes the full workflow from requirement research → research review → planning → plan review → development → development review using → test AI large language models. Capable of autonomously handling medium to large-scale engineering projects.
Command Claude Code
A command that takes a requirement through research, planning, plan review, development, and development review.
Command Claude Code
A command for finding the root causes of omissions or defects in work produced by the auto-work process.
Command Claude Code
A command for reviewing bug-fix code to check whether it fixes the underlying cause and introduces no side effects.
Command Claude Code
A command for fixing a bug and recording the lesson as a reusable skill or rule.
Command Claude Code
A command for recording a bug in the feature documentation for the relevant software version.
Command Claude Code
A command for making small code changes through implementation, compilation, and related tests.
Command Claude Code
A command for reviewing code produced during feature development, including checks for missed work and violations of project rules.
Command Claude Code
A command that repeatedly performs coding and review until the implementation reaches a stable quality level.
Command Claude Code
An automatic mode for implementing feature code, called by a feature-development loop, without asking the user questions during the process.
Command Claude Code
A guided development command that first parses its inputs, establishes context, confirms the implementation scope, and then follows a defined workflow.
Command Claude Code
An automatic version of the plan-creation command that builds context and makes requirement decisions without user interaction.
Command Claude Code
A command for creating an implementation plan together with an AI assistant from a set of requirements.
Command Claude Code
A command for reviewing an implementation plan, including its edge cases, covered scenarios, and architectural choices.
Command Claude Code
A command that repeatedly creates and reviews an implementation plan until the plan reaches a stable result.
Command Claude Code
A command for staging code changes and creating a conventionally formatted Git commit, which records a set of changes in the project history.
Command Claude Code
A command for sending the current code branch to a remote Git repository, which is the shared copy stored outside the local computer.
Command Claude Code
A development workflow with human checkpoints: people review the requirements and plan early, while later work runs independently.
Command Claude Code
A command for researching technical approaches by searching industry solutions and comparing them with project needs.
Command Claude Code
A command that repeatedly runs research and review steps until the results converge, meaning further rounds stop producing important changes.
Command Claude Code
A command for reviewing research reports. It checks the quality of the information, whether the proposed options cover the problem, and whether the conclusions make sense.
Command Claude Code
A command for starting, stopping, or restarting three development services: the backend, MCP service, and frontend. MCP is a service that lets an agent connect to tools or data.
Command Claude Code
A command for starting or stopping a GameServer ranking service. A ranking service records or serves an ordered list of game results or players.
Command Claude Code
A command for end-to-end smoke tests of agent calls. Smoke tests are quick checks that a system’s main paths still work.