implementation
Skill0 starsChanged 2 months agoMITView on GitHub
Guide TDD implementation. Red-green-refactor cycle, multi-layer testing, acceptance criteria at correct levels, Design by Contract, and version control discipline. Use when writing code for approved solutions.
16 other projects use a version of this.
Adopted by
Most recently taken up by
- thunderstormwang/AgentSkillstheir own version, first seen Nov 6, 2026
- Zejnilovic/agdtheir own version, first seen Nov 1, 2026
- jackstine/forgectltheir own version, first seen Nov 1, 2026
- JosefGisis/skillstheir own version, first seen Oct 20, 2026
- jvalin17/agent-toolkittheir own version, first seen Oct 12, 2026
- krishnapollu/my-devsquadtheir own version, first seen Oct 9, 2026
Used with
Projects with it also use bug-fix, investigate and 4 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
u9401066Skillbug-fix
Structured bug fixing workflow with root cause analysis and regression prevention. Triggers: BF, 修 bug, fix bug, debug, 除錯, 錯誤, error, 問題, issue, 修復, fix, 故障, fault, 異常, exception, 失敗, failure, 不能用, broken, 壞掉.
Also in 3 projects that use this
VercelSkillinvestigate
Investigate an issue report, or analyze the constraints of a proposed change, with strict, evidence-driven diagnosis.
Also in 3 projects that use this
MicrosoftSkillplaywright-cli
Automate browser interactions, test web pages and work with Playwright tests.
Also in 4 projects that use this
SentrySkillreview
Fixture review skill.
Also in 3 projects that use this
MicrosoftSkillrelease
Use ONLY when the user explicitly asks to cut/publish a Huabu desktop release or tag a version. Performs an outward action — bumps the version in apps/desktop/package.json, commits + annotated-tags vX.Y.Z, and pushes the branch + tag. That tag push auto-builds the macOS/Windows installers and publishes a GitHub Release. Provides `status`, `bump`, `tag-push`, and a one-command `run` (bump → tag-push).
Also in 4 projects that use this
SentrySkillskill-creator
Create new agent skills following the Agent Skills specification. Use when asked to "create a skill", "add a new skill", "write a skill", "make a skill", "build a skill", or scaffold a new skill with SKILL.md. Guides through requirements, writing, registration, and verification.
Also in 4 projects that use this
Other versions16 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 16
- bholmesdev/hubble.md★ 1.5ksame name
- ceedaragents/cyrus★ 888same name
- cyrusagents/cyrus★ 888same name
- EmeaAppGbb/spec2cloud★ 100same name
- jackstine/forgectl★ 8same name
- Zejnilovic/agd★ 5same name
- crgarcia12/Liliput★ 4same name
- websublime/mister-anderson★ 3same name
- jvalin17/agent-toolkit★ 2same name
- JosefGisis/skills★ 1same name
6 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository17 other files it gives its agents.
- analysis-reviewSkill
- business-analysisSkill
- change-assessmentSkill
- design-reviewSkill
- feedbackSkill
- installSkill
- operationsSkill
- project-definitionSkill
- project-setupSkill
- project-startSkill
- release-planningSkill
- retirementSkill
- risk-managementSkill
- roadmapSkill
- test-adequacy-reviewSkill
- testing-strategySkill
- validationSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
14 rules. 0 are common, 14 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 15 edited versions we read keep these.
14
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Use this skill when approved design work is ready to become working software.
- Distinctive: line 44
If the work does not trace cleanly, surface that before implementation proceeds.
- Distinctive: line 46
Create or reference the feature doc.
- Distinctive: line 48
If a feature doc does not exist, create one from the template.
- Distinctive: line 52
If the work is already story-sized, reference or create the appropriate story doc directly.
- Distinctive: line 62
Write it at the appropriate layer before implementing any enforcement point.
- Distinctive: line 63
Use inner-loop TDD (step 6) to drive each enforcement point's implementation.
- Distinctive: line 64
When the inner loop is green, the outer-loop test should pass.
- Distinctive: line 65
If it still fails, the gap is at the integration layer.
- Distinctive: line 67
Follow TDD for each story.
- Distinctive: line 73
If constraints force an expedient implementation, log the gap between the expedient and the ideal.
- Distinctive: line 81
Keep this as a structured conversational gate unless a project proves that a persistent artifact is needed.
- Distinctive: line 95
All tests pass at every layer defined in the testing strategy
- Distinctive: line 109
Use the templates in this skill directory when scaffolding or substantially reshaping those artifacts.
“Common” is measured: kept by at least 6 in 10 of the 15 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Implementation
Use this skill when approved design work is ready to become working software.
This skill assumes the change has already passed through the earlier lifecycle gates. It is for disciplined delivery, not for skipping analysis or design.
Current Project Context
Approved Business Analysis
!cat docs/02-analysis/business-analysis.md 2>/dev/null || echo "No business analysis found. Implementation should not proceed for non-trivial work without analysis."
Requirements
!cat docs/02-analysis/requirements/requirements.md 2>/dev/null || echo "No requirements found."
Design Review Status
!cat docs/03-design/design-review-checklist.md 2>/dev/null || echo "No design review checklist found. Confirm design approval before implementation."
Design Traceability
!cat docs/03-design/04-traceability-matrix.md 2>/dev/null || echo "No design traceability matrix found."
Testing Strategy
!cat docs/03-design/05-testing-strategy.md 2>/dev/null || echo "No testing strategy found. Use the project testing approach if defined elsewhere, or note the gap explicitly."
Existing Implementation Artifacts
!find docs/04-implementation/features -name "*.md" | sort 2>/dev/null || echo "No feature or story docs found."
Active Technical Debt
!cat docs/04-implementation/technical-debt.md 2>/dev/null || echo "No technical debt log found."
Interaction Pattern
- Confirm traceability before writing code.
- Identify which business analysis solution(s), requirement(s), and design element(s) this work addresses.
- If the work does not trace cleanly, surface that before implementation proceeds.
- Create or reference the feature doc.
- Feature docs carry feature-level acceptance criteria for cross-story or integration behaviors.
- If a feature doc does not exist, create one from the template.
- Break the feature into stories when needed.
- Story docs carry single-behavior, single-component acceptance criteria.
- If the work is already story-sized, reference or create the appropriate story doc directly.
- Classify each acceptance criterion before writing any test.
- Draw criteria from the User Manual's procedure steps as well as from the requirements. A step carrying a state precondition ("the year file exists and is included before the year's first import") is a criterion, and it spans more than one component by construction. Weight the obligation toward rare steps: a daily step is exercised into correctness for free, while an annual one has a twelve-month feedback loop that a millisecond test replaces.
- Enumerate the criterion's enforcement points: every component that must independently uphold it for the criterion to hold from the user's perspective.
- One enforcement point: story-level criterion; inner-loop TDD alone is sufficient.
- More than one: feature-level criterion; an outer-loop test is required. A unit test on one enforcement point is not sufficient coverage.
- Rule of thumb: "the user sees X" is always feature-level — what the user sees is the product of every layer between the data and the screen.
- For feature-level criteria, write the failing outer-loop test first.
- Write it at the appropriate layer before implementing any enforcement point.
- Use inner-loop TDD (step 6) to drive each enforcement point's implementation.
- When the inner loop is green, the outer-loop test should pass.
- If it still fails, the gap is at the integration layer.
- Follow TDD for each story.
- Red: write the minimal failing test from an acceptance criterion. Run it and confirm failure. Do not write implementation yet.
- Green: write the simplest code that makes the test pass. No more.
- Refactor: improve structure without changing behavior. All tests must still pass.
- Record technical debt intentionally.
- If constraints force an expedient implementation, log the gap between the expedient and the ideal.
- Tie debt to the relevant feature or story and include a repayment trigger.
- Present an implementation checkpoint at meaningful milestones.
- Summarize what was built.
- Report test results.
- Confirm traceability.
- Name debt, blockers, or open issues.
- Keep this as a structured conversational gate unless a project proves that a persistent artifact is needed.
Anti-Patterns To Flag
- Writing tests and implementation simultaneously
- Tautological tests
- Skipping the refactor step
- Batching multiple behaviors into one test
- Integration criteria placed at story level
- Component criteria placed at feature level
- Declaring a feature-level criterion covered by unit tests on one enforcement point
Definition Of Done
- All tests pass at every layer defined in the testing strategy
- Code passes automated review
- The feature traces back to business analysis solution(s)
- Documentation is updated when applicable
- Technical debt, if any, is recorded intentionally
Outputs
Produce or update these artifacts as needed:
docs/04-implementation/features/feature-*.mddocs/04-implementation/features/stories/story-*.mddocs/04-implementation/technical-debt.md
Use the templates in this skill directory when scaffolding or substantially reshaping those artifacts.
Gate
Human review is required at meaningful implementation checkpoints before treating a feature milestone as accepted.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

