Personal agent skills bundle for AI-assisted software development — grill, plan, implement, and ship with hard quality gates. Install once, use in any codebase across AI IDEs.
Agent
An architecture agent that explores a problem and produces design decisions for later implementation work. It focuses on understanding the domain, constraints, modules, interfaces, and alternatives before code is written.
Agent
An implementation agent that turns a design and task breakdown into code and tests. TDD, or test-driven development, means writing a failing test first, then the smallest code that makes it pass.
Agent
A code-review agent that examines proposed changes from an independent point of view. It compares the code with design notes, technical specifications, and mockups, then records findings without editing files.
Agent
A design-review agent that examines an architecture proposal against the domain model and existing architecture decisions. It reports problems without changing the design documents.
Agent
A Git operations agent that handles routine version-control work such as synchronising a repository, creating commits, and pushing changes. It does not edit code files.
Agent
A planning agent that turns an approved design into an ordered set of implementation tasks. It creates product requirements, an implementation plan, and individual issue files.
Agent
An agent for quickly building small prototypes to test whether an idea can work. It prioritizes speed and a clear feasibility verdict over production-quality code.
Agent
A quality-audit agent that looks specifically for overengineering: extra abstractions, premature optimisation, and work outside the agreed scope. It does not review whether the code is correct.
Agent
A Chinese-language interface-design agent that turns product requirements into previewable HTML mockups showing how screens should look.