Wukong Code

Make coding agents work like engineers

A software development methodology for coding agents.

Discovery, design, TDD, review, verification. Beyond generating code, finish the engineering.

26 top-level skills · 10 development environments · zero core dependencies · MIT licensed

  1. Clarify goals, boundaries, and acceptance criteria

  2. Compare approaches and obtain approval

  3. Break down tasks, risks, and verification commands

  4. RED → GREEN → REFACTOR

    • RED
    • GREEN
    • REFACTOR
  5. Check requirement fit and code quality

    evidence attached
  6. Run fresh tests, builds, and reproductions

    PASS
  7. You choose the final integration path

A failed check returns work to the stage that needs evidence

The problem

AI writes fast. Reliable delivery is another matter.

  1. The code is generated before the requirement is clear.

  2. The bug is not located, so every change is a guess.

  3. The agent declares success before the tests run.

  4. The missing engineering process: design, tests, verification.

Core capabilities

Reliable engineering, by default

  1. Design before coding

    Clarify the goal, compare approaches, then start.

  2. Real test-driven development

    See the failure first, then write the smallest implementation.

  3. Systematic debugging

    Reproduce, gather evidence, find the root cause, then fix.

  4. Review throughout

    Keep checking requirement fit and code quality.

  5. Verify before declaring done

    Without fresh tests or builds, it is not done.

  6. Triggered by the task

    Skills pick the workflow from context. No repeated prompts.

Delivery evidence

From idea to reliable delivery

Simple tasks stay simple. Complex tasks get constraints.

$ wukong run --task feature

  1. discovery ............. ✓
  2. design approval ....... ✓
  3. implementation plan ... ✓
  4. TDD (RED) ............. ✓
  5. TDD (GREEN) ........... ✓
  6. code review ........... ✓ evidence attached
  7. final verification .... ✓ PASS
  8. merge / PR ............ ready

Product design

Product Design, same workflow

From research, audits, and visual exploration to implementation and Design QA.

  1. 01 · VISUAL EXPLORE Choose a direction
  2. 02 · BUILD Implement the real interface
  3. 03 · DESIGN QA Compare the rendered result
Visual target
Delivery workflow
Design → BuildDesign QA

Project dashboard

  • Requirements review To organize
  • Interaction states Unspecified
  • Design checks To review
Rendered result
Delivery workflow
Design → BuildDesign QA

Project dashboard

  • Requirements review Verified
  • Interaction states Covered
  • Design checks Passed

Method comparison

More than a prompt template

How ordinary prompts and Wukong Code approach software work
Dimension Ordinary prompt Wukong Code
Trigger Depends on manual reminders Triggers from the task
Scope Focuses on code generation Covers the delivery workflow
Debugging Relies on guesses Investigates the root cause first
Completion Depends on judgment Requires verification evidence

Environments

Use it where you already work

Zero core dependencies. No extra tools. No overwritten commands.

  1. Antigravity
  2. Claude Code
  3. Codex App
  4. Codex CLI
  5. Cursor
  6. Factory Droid
  7. GitHub Copilot CLI
  8. Kimi Code
  9. OpenCode
  10. Pi

Next step

Code can be fast. Quality cannot depend on luck.

Let agents do the right work at the right time.

More than code—delivery you can trust.