Real engineering with AI
A disciplined core for alignment, architecture, implementation, tests, diagnosis, and review.
Why this collection exists
Keep the speed of coding agents without surrendering engineering judgment or feedback loops.
Prompts
Use these directly when the moment calls for a focused process.
Grill my feature idea
Expose unclear requirements and hidden decisions before implementation begins.
Map an unfamiliar codebase
Build a useful model of modules, data flow, boundaries, and domain language before changing code.
Implement one vertical slice
Deliver a small end-to-end behavior with tests and repository-aligned structure.
Review an implementation before calling it finished
Audit a completed change against intent, repository standards, and forgotten states.
Skills
Install these when the discipline should remain available across sessions.
Grill With Docs
Sharpens a plan while building durable domain language and decision records in the repository.
Test-Driven Development
Runs a focused red–green–refactor loop one vertical behavior at a time.
Diagnosing Bugs
Uses reproduction, minimization, hypotheses, and instrumentation before implementing a fix.
Code Review
Reviews a change independently for engineering standards and specification fidelity.
Workflows
Follow these when the work spans several distinct stages.
From rough idea to buildable specification
Resolve the important product decisions, cut excess scope, and leave a specification that implementation can follow.
Ship one vertical feature slice
Implement a settled behavior through the existing architecture with a short feedback loop and controlled scope.
Review a pull request thoroughly
Review the change on two axes—engineering quality and fidelity to the promised behavior.