feature-planner
Skill14 starsChanged 37 days agoMITView on GitHub
Phase-based feature planning with quality gates and incremental delivery structure. Use when breaking down features, organizing work, or structuring roadmaps.
9 other projects use a version of this, 2 of them edited.
Adopted by
Most recently taken up by
- vincenzo345/skillstheir own version, first seen Nov 8, 2026
- ralvarez-ai/ralvaskillstheir own version, first seen Nov 1, 2026
- wonhp1/claude-skillsedited, first seen Oct 20, 2026
- DMcSweeney/ClaudeSkillstheir own version, first seen Oct 15, 2026
- serendipity1004/cc-feature-implementeredited, first seen Oct 8, 2026
- coinbase/cdstheir own version, first seen Oct 8, 2026
Used with
Projects with it also use improve-codebase-architecture.
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.
GoogleSkillimprove-codebase-architecture
Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.
Also in 3 projects that use this
Other versions9 projects use a version of this.
Word for word · 0
None found.
Edited · 2
- serendipity1004/cc-feature-implementer★ 17697% same
- wonhp1/claude-skills★ 075% same
Wrote their own version · 7
- coinbase/cds★ 509same name
- ralvarez-ai/ralvaskills★ 4same name
- ralvarezdev/ralvaskills★ 4same name
- Su-Yong/skills★ 4same name
- vincenzo345/skills★ 1same name
- DMcSweeney/ClaudeSkills★ 0same name
- ubin8/Lovable-Cookbook★ 0same name
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 repository44 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- adaptSkill
- animateSkill
- auditSkill
- bolderSkill
- brainstormSkill
- build-fixSkill
- business-panelSkill
- checkpointSkill
- clarifySkill
- colorizeSkill
- composition-patternsSkill
- confidence-checkSkill
- config-doctorSkill
- critiqueSkill
- delightSkill
- design-auditSkill
- devils-advocateSkill
- distillSkill
- eval-harnessSkill
- fix-prSkill
- frontend-designSkill
- gap-analysisSkill
- grill-with-docsSkill
- hardenSkill
- impeccableSkill
- layoutSkill
- learnSkill
- optimizeSkill
- overdriveSkill
- polishSkill
- pytest-runnerSkill
- python-best-practicesSkill
- quieterSkill
- react-best-practicesSkill
- sc-helpSkill
- security-auditSkill
- shapeSkill
- tddSkill
- typesetSkill
- ui-ux-pro-maxSkill
- uv-packageSkill
- verifySkill
- web-design-guidelinesSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
20 rules. 0 are common, 18 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 8 edited versions we read keep these.
18
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 19
Read relevant files to understand codebase architecture
- Distinctive: line 26
Test-First: Write tests BEFORE implementation
- Distinctive: line 39
RED Tasks: Write failing tests first
- Very specific: line 47
Use plan-template.md to generate: docs/plans/PLAN_<feature-name>.md
- Distinctive: line 60
CRITICAL: Use AskUserQuestion to get explicit approval before proceeding.
- Distinctive: line 67
Only create plan document after user confirms approval.
- Very specific: line 70
Create docs/plans/ directory if not exists
- Distinctive: line 72
Add clear instructions in header about quality gates
- Distinctive: line 116
Add this to plan document header:
- Distinctive: line 156
Quality Risks: Test coverage gaps, regression potential
- Distinctive: line 184
Write tests that WILL fail
- Distinctive: line 185
Verify tests fail for the right reason
- Distinctive: line 186
Run tests to confirm failure
- Distinctive: line 187
Commit failing tests to track TDD compliance
- Distinctive: line 190
Write minimal code to make tests pass
- Distinctive: line 191
Run tests frequently (every 2-5 minutes)
- Distinctive: line 192
Stop when all tests pass
- Distinctive: line 193
No additional functionality beyond tests
- Distinctive: line 199
Run tests after each refactoring step
- Distinctive: line 200
Commit when refactoring complete
“Common” is measured: kept by at least 6 in 10 of the 8 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
Feature Planner
Purpose
Generate structured, phase-based plans where:
- Each phase delivers complete, runnable functionality
- Quality gates enforce validation before proceeding
- User approves plan before any work begins
- Progress tracked via markdown checkboxes
- Each phase is 1-4 hours maximum
Planning Workflow
Step 1: Requirements Analysis
- Read relevant files to understand codebase architecture
- Identify dependencies and integration points
- Assess complexity and risks
- Determine appropriate scope (small/medium/large)
Step 2: Phase Breakdown with TDD Integration
Break feature into 3-7 phases where each phase:
- Test-First: Write tests BEFORE implementation
- Delivers working, testable functionality
- Takes 1-4 hours maximum
- Follows Red-Green-Refactor cycle
- Has measurable test coverage requirements
- Can be rolled back independently
- Has clear success criteria
Phase Structure:
- Phase Name: Clear deliverable
- Goal: What working functionality this produces
- Test Strategy: What test types, coverage target, test scenarios
- Tasks (ordered by TDD workflow):
- RED Tasks: Write failing tests first
- GREEN Tasks: Implement minimal code to make tests pass
- REFACTOR Tasks: Improve code quality while tests stay green
- Quality Gate: TDD compliance + validation criteria
- Dependencies: What must exist before starting
- Coverage Target: Specific percentage or checklist for this phase
Step 3: Plan Document Creation
Use plan-template.md to generate: docs/plans/PLAN_<feature-name>.md
Include:
- Overview and objectives
- Architecture decisions with rationale
- Complete phase breakdown with checkboxes
- Quality gate checklists
- Risk assessment table
- Rollback strategy per phase
- Progress tracking section
- Notes & learnings area
Step 4: User Approval
CRITICAL: Use AskUserQuestion to get explicit approval before proceeding.
Ask:
- "Does this phase breakdown make sense for your project?"
- "Any concerns about the proposed approach?"
- "Should I proceed with creating the plan document?"
Only create plan document after user confirms approval.
Step 5: Document Generation
- Create
docs/plans/directory if not exists - Generate plan document with all checkboxes unchecked
- Add clear instructions in header about quality gates
- Inform user of plan location and next steps
Quality Gate Standards
Each phase MUST validate these items before proceeding to next phase:
Build & Compilation:
- Project builds/compiles without errors
- No syntax errors
Test-Driven Development (TDD):
- Tests written BEFORE production code
- Red-Green-Refactor cycle followed
- Unit tests: ≥80% coverage for business logic
- Integration tests: Critical user flows validated
- Test suite runs in acceptable time (<5 minutes)
Testing:
- All existing tests pass
- New tests added for new functionality
- Test coverage maintained or improved
Code Quality:
- Linting passes with no errors
- Type checking passes (if applicable)
- Code formatting consistent
Functionality:
- Manual testing confirms feature works
- No regressions in existing functionality
- Edge cases tested
Security & Performance:
- No new security vulnerabilities
- No performance degradation
- Resource usage acceptable
Documentation:
- Code comments updated
- Documentation reflects changes
Progress Tracking Protocol
Add this to plan document header:
**CRITICAL INSTRUCTIONS**: After completing each phase:
1. ✅ Check off completed task checkboxes
2. 🧪 Run all quality gate validation commands
3. ⚠️ Verify ALL quality gate items pass
4. 📅 Update "Last Updated" date
5. 📝 Document learnings in Notes section
6. ➡️ Only then proceed to next phase
⛔ DO NOT skip quality gates or proceed with failing checksPhase Sizing Guidelines
Small Scope (2-3 phases, 3-6 hours total):
- Single component or simple feature
- Minimal dependencies
- Clear requirements
- Example: Add dark mode toggle, create new form component
Medium Scope (4-5 phases, 8-15 hours total):
- Multiple components or moderate feature
- Some integration complexity
- Database changes or API work
- Example: User authentication system, search functionality
Large Scope (6-7 phases, 15-25 hours total):
- Complex feature spanning multiple areas
- Significant architectural impact
- Multiple integrations
- Example: AI-powered search with embeddings, real-time collaboration
Risk Assessment
Identify and document:
- Technical Risks: API changes, performance issues, data migration
- Dependency Risks: External library updates, third-party service availability
- Timeline Risks: Complexity unknowns, blocking dependencies
- Quality Risks: Test coverage gaps, regression potential
For each risk, specify:
- Probability: Low/Medium/High
- Impact: Low/Medium/High
- Mitigation Strategy: Specific action steps
Rollback Strategy
For each phase, document how to revert changes if issues arise. Consider:
- What code changes need to be undone
- Database migrations to reverse (if applicable)
- Configuration changes to restore
- Dependencies to remove
Test Specification Guidelines
Test-First Development Workflow
For Each Feature Component:
- Specify Test Cases (before writing ANY code)
- What inputs will be tested?
- What outputs are expected?
- What edge cases must be handled?
- What error conditions should be tested?
- Write Tests (Red Phase)
- Write tests that WILL fail
- Verify tests fail for the right reason
- Run tests to confirm failure
- Commit failing tests to track TDD compliance
- Implement Code (Green Phase)
- Write minimal code to make tests pass
- Run tests frequently (every 2-5 minutes)
- Stop when all tests pass
- No additional functionality beyond tests
- Refactor (Blue Phase)
- Improve code quality while tests remain green
- Extract duplicated logic
- Improve naming and structure
- Run tests after each refactoring step
- Commit when refactoring complete
Test Types
Unit Tests:
- Target: Individual functions, methods, classes
- Dependencies: None or mocked/stubbed
- Speed: Fast (<100ms per test)
- Isolation: Complete isolation from external systems
- Coverage: ≥80% of business logic
Integration Tests:
- Target: Interaction between components/modules
- Dependencies: May use real dependencies
- Speed: Moderate (<1s per test)
- Isolation: Tests component boundaries
- Coverage: Critical integration points
End-to-End (E2E) Tests:
- Target: Complete user workflows
- Dependencies: Real or near-real environment
- Speed: Slow (seconds to minutes)
- Isolation: Full system integration
- Coverage: Critical user journeys
Test Coverage Calculation
Coverage Thresholds (adjust for your project):
- Business Logic: ≥90% (critical code paths)
- Data Access Layer: ≥80% (repositories, DAOs)
- API/Controller Layer: ≥70% (endpoints)
- UI/Presentation: Integration tests preferred over coverage
Coverage Commands by Ecosystem:
# JavaScript/TypeScript
jest --coverage
nyc report --reporter=html
# Python
pytest --cov=src --cov-report=html
coverage report
# Java
mvn jacoco:report
gradle jacocoTestReport
# Go
go test -cover ./...
go tool cover -html=coverage.out
# .NET
dotnet test /p:CollectCoverage=true /p:CoverageReporter=html
reportgenerator -reports:coverage.xml -targetdir:coverage
# Ruby
bundle exec rspec --coverage
open coverage/index.html
# PHP
phpunit --coverage-html coverageCommon Test Patterns
Arrange-Act-Assert (AAA) Pattern:
test 'description of behavior':
// Arrange: Set up test data and dependencies
input = createTestData()
// Act: Execute the behavior being tested
result = systemUnderTest.method(input)
// Assert: Verify expected outcome
assert result == expectedOutputGiven-When-Then (BDD Style):
test 'feature should behave in specific way':
// Given: Initial context/state
given userIsLoggedIn()
// When: Action occurs
when userClicksButton()
// Then: Observable outcome
then shouldSeeConfirmation()Mocking/Stubbing Dependencies:
test 'component should call dependency':
// Create mock/stub
mockService = createMock(ExternalService)
component = new Component(mockService)
// Configure mock behavior
when(mockService.method()).thenReturn(expectedData)
// Execute and verify
component.execute()
verify(mockService.method()).calledOnce()Test Documentation in Plan
In each phase, specify:
- Test File Location: Exact path where tests will be written
- Test Scenarios: List of specific test cases
- Expected Failures: What error should tests show initially?
- Coverage Target: Percentage for this phase
- Dependencies to Mock: What needs mocking/stubbing?
- Test Data: What fixtures/factories are needed?
Supporting Files Reference
- plan-template.md - Complete plan document template
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.

