A command for reviewing learned instincts, meaning recurring patterns an agent has recorded from past work, and proposing new skills, commands, or agents. It can also generate files from those proposals.
A Go development command for finding and fixing build, compilation, and vet errors. Go is a programming language, and vet is a tool that spots likely mistakes in Go code.
A code-review command for Go, the programming language, that checks common Go patterns, error handling, concurrency, performance, and code organization.
A command for exporting learned coding instincts, meaning reusable rules about how work should be handled, from a continuous-learning system. It saves them as a shareable JSON file.
A command for importing instincts, meaning small learned rules about how coding work should be handled, into a continuous-learning system. It accepts instincts from a file, URL, or teammate source.
A command that displays learned instincts, meaning reusable rules gathered by a continuous-learning system, together with their confidence levels. It shows both project-specific and global rules.
A command for finding unused code and combining duplicated code in a project. Dead code is code that is no longer used, while duplicated code repeats the same logic in multiple places.
A command for reviewing code for common security weaknesses, using the OWASP checklist. OWASP is a group that publishes widely used guidance for web application security.
A command for choosing which package manager a project should use, such as npm, pnpm, Yarn, or Bun. A package manager installs project libraries and runs project scripts.
A command that applies test-driven development (TDD), where you write a failing test before the code, then make it pass and improve the code. It targets at least 80% test coverage, with stricter targets for critical logic.
A command for measuring which parts of a codebase are exercised by tests. Test coverage shows the proportion of statements, branches, functions, or lines that tests run.
A command that creates or refreshes code maps: documents showing a codebase's structure, important files, modules, public interfaces, and relationships.
A verification command that runs checks such as type checking, linting, tests, coverage, security review, and building the project. Type checking checks code types, while linting checks common code-quality problems.
A Chinese-language CLAUDE.md instruction file for a Claude Code plugin, which is a collection of AI coding tools and instructions. It documents the project structure, testing commands, architecture, and commands such as TDD, planning, end-to-end testing, and code review; TDD means writing tests to guide the implementation.
A software-architecture advisor for planning systems, designing new features, and making technical decisions. Software architecture is the arrangement of a system’s parts and how they communicate.