A skill for completing university assignments that contain several kinds of work, such as a report and a program. It divides the requirements between the appropriate skills and checks that the finished pieces agree with each other.
A workflow for implementing the code portion of a programming task from its specification and grading rubric. It includes checking the requirements, testing the result, and preparing the code for final refinement.
A writing skill for reports, essays, and answers to university assignment questions. It uses the assignment instructions and grading criteria to guide the document, then checks its structure, sources, facts, and required content.
A project-wide refinement process that reviews different kinds of project files, delegates improvements to matching tools, and then checks consistency between them.
A release guide for publishing a software package, including projects with multiple independently versioned packages in one repository, called a monorepo. It determines the release unit, updates versions using Semantic Versioning, and prepares the related GitHub release steps after approval.
A code-review workflow that checks whether an implementation matches the project's agreed plan and evaluation criteria. It uses `.claude/context.md` as the reference document.
A project-discovery workflow that creates `.claude/context.md`, a shared file describing the project's purpose, design, constraints, and measures of success.
A discussion workflow for shaping a project's theory, hypotheses, and overall direction. It records agreed decisions in `.claude/context.md`, a project context file.
A general repair skill for when an add-on gives a wrong result, misses something, or receives negative feedback. It finds the rule or instruction that caused the problem and updates it.
A cleanup workflow for removing Claude Code's local history, temporary traces, and caches while preserving login details, settings, and installed plugins.
A procedure for running programs that may take more than a few dozen seconds, such as training jobs, evaluations, data processing, or servers, while recording logs and detecting failures.
A code-maintenance guide written in Japanese for polishing scripts, configuration, and other project code while keeping its externally visible behavior unchanged.