Incremental Implementation
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md and install @user_3c6cb52e/incremental-implementation-lr-x2ll.
About this skill
Problem
Multi-file changes and new features often fail when too much code is written at once: errors are harder to localize, tests fail late, and commits become noisy. SKILL.md focuses on avoiding one-shot feature implementation and instead advancing through verifiable increments.
How It Works
The skill frames implementation as small slices that keep the system working and testable:
- Implement: build the smallest complete piece of functionality instead of laying down large code first
- Test: run existing tests or add a test for the slice
- Verify: confirm
build,tests, and manual checks pass - Commit: save progress with a descriptive message
- Continue: carry the current state into the next slice without restarting
It favors vertical slices, such as one complete path through API, service, and UI, before expanding the scope.
Boundaries
Use it for multi-file changes, task-based feature work, refactoring, and especially when you are about to write more than roughly 100 lines before testing. For small single-file or single-function changes, skipping the process can be reasonable.
Use Cases
- Refactor legacy modules by making the smallest change first, running tests, and avoiding replacing entire files at once.
- Build a new feature by completing one end-to-end request path first, then expanding other branches.
- Before changing over a hundred lines, split work into a smallest verifiable unit and add tests.
- During multi-file task breakdowns, confirm each slice with test and build results before continuing.
Best For
- Full-stack engineers implementing frontend-service features who need to avoid large untested code changes.
- Backend engineers refactoring legacy code who need small, testable changes.
- Senior engineers delivering task-based features who need to break work into committable slices.
- Tech leads guiding AI assistants who need a steady implement-test-verify-commit rhythm.
Related Skills
Local workflow memory with matching and SOP updates.
An OpenClaw live streaming executor that initializes TRTC streaming, starts a real-time dashboard, generates viewer URLs, and continuously reports live events.
Breaks down physical supply chains for super-trends to identify second- and third-layer bottlenecks, runs valuation and reverse checks, and maintains trackable reports.
A token-saving compression mode for Chinese LLMs with lite, full, ultra, and classical tiers, preserving code and technical terms while handling edge cases.