plugin

Software Engineering Toolkit

Ship disciplined software

Software Engineering Toolkit is an adapted multi-skill engineering toolkit inspired by Matt Pocock's open-source skills repository. It helps agents and developers align on intent, model the domain, write specs and tickets, design deep modules, use test-driven development, diagnose bugs, review code, research with citations, create handoffs, and improve codebase architecture.

Install

  1. 1Add it in ChatGPT

    Open Plugins in the ChatGPT sidebar (web or desktop), search for “Software Engineering Toolkit” and select +. You'll be asked to sign in to any connected service on first use. Then type @ in a chat to call it.

Skills 15

  • code-review

    Review changes against both engineering standards and the originating specification.

  • codebase-design

    Design deep modules with small interfaces, clear seams, and testable responsibilities.

  • diagnosing-bugs

    Diagnose bugs and regressions through reproduction, minimization, instrumentation, fixing, and regression testing.

  • domain-modeling

    Build a shared vocabulary and test it against real scenarios before changing code.

  • grilling

    Interview the user until a plan, design, or decision is specific enough to execute.

  • handoff-and-context

    Create compact handoffs and shared project context so another agent can continue accurately.

  • implementation

    Implement specs or tickets in small slices with tests, review, and a clear completion report.

  • merge-conflicts

    Resolve merge or rebase conflicts by tracing intent and preserving both sides where appropriate.

  • prototyping

    Build small throwaway prototypes to answer uncertain design, interaction, or technical questions.

  • research

    Investigate engineering questions with high-trust primary sources and cited findings.

  • software-engineering-toolkit

    Route engineering work into disciplined alignment, design, testing, diagnosis, review, research, and delivery workflows.

  • tdd

    Drive implementation with a red-green-refactor loop and meaningful behavior tests.

  • to-spec

    Turn a conversation or request into a precise, testable engineering specification.

  • to-tickets

    Break a plan into small tracer-bullet tickets with dependencies and completion criteria.

  • writing-for-agents

    Write concise, discoverable instructions and project documents that agents can follow reliably.

Things to ask

  • “Grill me until this engineering plan is unambiguous.”
  • “Turn this request into a testable implementation spec.”
  • “Diagnose this bug with a disciplined feedback loop.”

Related