agentleFS
TehseenTechTehseenTech/manus-skills/skills/workflow-skills/systematic-feature-builder/SKILL.md

systematic-feature-builder

Skill0 starsChanged 3 months agoMITView on GitHub

Systematic approach to implementing features with detailed planning, tracking, and testing. Use when building new features, adding functionality to existing projects, implementing multi-step requirements, or when user requests organized, methodical development workflow. Apply automatically when the request matches; the user does not need to mention this skill.

1 other project uses a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

26 rules. About 0 are common, 23 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈23

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when the request matches its description.

    line 15
  2. Distinctive:

    Use this skill when:

    line 23
  3. Distinctive:

    Follow these five phases in order:

    line 33
  4. Distinctive:

    Implementation Planning: - Create detailed, step-by-step plan

    line 36
  5. Distinctive:

    Documentation & Checkpoint: - Create comprehensive checkpoint

    line 39
  6. Distinctive:

    Ask targeted questions if requirements are unclear:

    line 61
  7. Distinctive:

    Use the implementation plan template:

    line 71
  8. Distinctive:

    Add all planned tasks to project's todo.md:

    line 118
  9. Very specific:

    Use file tool's edit action to update checkboxes efficiently.

    line 156
  10. Distinctive:

    If you deviate from the plan:

    line 160
  11. Distinctive:

    Document why in the plan file

    line 161
  12. Distinctive:

    Update todo.md with new tasks

    line 162
  13. Distinctive:

    Test at all three levels before checkpoint:

    line 176
  14. Distinctive:

    Before creating checkpoint, verify all tasks complete:

    line 215
  15. Distinctive:

    Ensure documentation reflects changes:

    line 228
  16. Distinctive:

    Update README if user-facing features added

    line 229
  17. Distinctive:

    Add inline code comments for complex logic

    line 230
  18. Distinctive:

    Update API documentation if endpoints changed

    line 231
  19. Very specific:

    Follow checkpoint guidelines in references/checkpoint-guidelines.md.

    line 235
  20. Distinctive:

    Before creating webdev checkpoint:

    line 270
  21. Distinctive:

    Always follow this sequence:

    line 283
  22. Very specific:

    Run pnpm db:push to apply migration

    line 286
  23. Distinctive:

    Create files in this order:

    line 292
  24. Distinctive:

    Test → Verify fix + no regressions

    line 337
  25. Distinctive:

    Update todo.md after each task

    line 420
  26. Distinctive:

    Test each component independently

    line 421

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Overview

Systematic approach to implementing features with detailed planning, tracking, and testing. Use when building new features, adding functionality to existing projects, implementing multi-step requirements, or when user requests organized, methodical development workflow. Apply automatically when the request matches; the user does not need to mention this skill.

This skill should apply automatically when the request matches; the user does not need to mention this skill.

Usage

Use this skill when the request matches its description. The skill is discovered automatically; the user does not need to mention the skill explicitly.

Systematic Feature Builder

Implement features systematically with detailed planning, progress tracking, and comprehensive testing.

When to Use This Skill

Use this skill when:

  • Building new features from requirements or specifications
  • Adding functionality to existing projects
  • Implementing multi-step features that require coordination
  • User requests "systematic", "organized", or "methodical" approach
  • User provides a spec or detailed requirements document
  • Features require database changes, API integration, or complex logic

Core Workflow

Follow these five phases in order:

  1. Requirements Analysis - Understand what needs to be built
  2. Implementation Planning - Create detailed, step-by-step plan
  3. Systematic Implementation - Execute plan with progress tracking
  4. Testing & Verification - Validate at multiple levels
  5. Documentation & Checkpoint - Create comprehensive checkpoint

Phase 1: Requirements Analysis

Gather Information

Extract clear requirements from user input:

  • What features need to be built?
  • What is the expected behavior?
  • Are there existing examples or references?
  • What are the success criteria?

Identify Dependencies

Determine what exists and what needs to be created:

  • Existing code/components that can be reused
  • Database schema changes required
  • External services or APIs needed
  • Dependencies between features

Clarify Ambiguities

Ask targeted questions if requirements are unclear:

  • Edge cases and error handling
  • User permissions and access control
  • Performance or scalability requirements
  • UI/UX preferences

Phase 2: Implementation Planning

Create Detailed Plan

Use the implementation plan template:

# Copy template to project
cp /home/ubuntu/skills/systematic-feature-builder/templates/implementation-plan-template.md \
   <project-path>/docs/plans/YYYY-MM-DD-feature-name.md

Plan Structure:

  • Overview and objectives
  • Task breakdown with estimated time
  • Step-by-step instructions with file paths
  • Testing strategy for each task
  • Success criteria
  • Rollback plan

Break Down Into Atomic Tasks

Each task should:

  • Be completable in one sitting (30-90 minutes)
  • Have clear inputs and outputs
  • Be independently testable
  • Have explicit success criteria

Example Task Breakdown:

### Task 1: Add Database Field (15 min)
1. Update schema.ts with new field
2. Run `pnpm db:push` to migrate
3. Verify field exists in database

### Task 2: Create Backend Endpoint (30 min)
1. Add endpoint to router
2. Implement business logic
3. Add input validation
4. Write unit tests

### Task 3: Build Frontend Component (45 min)
1. Create component file
2. Add tRPC integration
3. Implement UI with shadcn/ui
4. Handle loading/error states

Update todo.md

Add all planned tasks to project's todo.md:

## [Feature Name]

- [ ] Task 1: Database changes
- [ ] Task 2: Backend endpoint
- [ ] Task 3: Frontend component
- [ ] Task 4: Testing
- [ ] Task 5: Documentation

Phase 3: Systematic Implementation

Execute Plan Sequentially

Implement tasks in order, following the plan:

  1. Start with backend/data layer (database, APIs)
  2. Then business logic (endpoints, services)
  3. Finally UI layer (components, pages)

This order minimizes rework and allows incremental testing.

Track Progress in todo.md

Mark tasks complete as you finish them:

## [Feature Name]

- [x] Task 1: Database changes
- [x] Task 2: Backend endpoint
- [ ] Task 3: Frontend component  ← Currently working on this
- [ ] Task 4: Testing
- [ ] Task 5: Documentation

Use file tool's edit action to update checkboxes efficiently.

Handle Deviations

If you deviate from the plan:

  • Document why in the plan file
  • Update todo.md with new tasks
  • Adjust time estimates if needed

Communicate Progress

For long implementations, send info messages at milestones:

  • "✅ Task 1 complete: Database schema updated"
  • "📊 Progress: 3/5 tasks complete"
  • "⚠️ Encountered issue with X, implementing alternative approach"

Phase 4: Testing & Verification

Three-Level Testing

Test at all three levels before checkpoint:

1. Unit Tests

  • Individual functions and endpoints
  • Database operations
  • Input validation

2. Integration Tests

  • Features working together
  • End-to-end workflows
  • External service integration

3. Manual Verification

  • UI renders correctly
  • User flows work as expected
  • Edge cases handled gracefully

Webdev-Specific Testing

For webdev projects:

# Run automated tests
pnpm test

# Check dev server health
# (use webdev_check_status tool)

# Manual browser testing
# - Open dev server URL
# - Test feature in browser
# - Check console for errors
# - Verify responsive design

Refer to references/testing-strategies.md for detailed testing approaches.

Verify Completion

Before creating checkpoint, verify all tasks complete:

python /home/ubuntu/skills/systematic-feature-builder/scripts/verify_completion.py \
  <project-path>/todo.md

This script checks for incomplete [ ] items and reports them.

Phase 5: Documentation & Checkpoint

Update Documentation

Ensure documentation reflects changes:

  • Update README if user-facing features added
  • Add inline code comments for complex logic
  • Update API documentation if endpoints changed

Create Comprehensive Checkpoint

Follow checkpoint guidelines in references/checkpoint-guidelines.md.

Checkpoint Description Template:

[Feature name]: [brief summary]

## Features Added

### 1. [Feature Name]
- Key capability 1
- Key capability 2
- Location/integration point

## Technical Details

**Database Changes:**
- [Schema changes]

**New Files:**
- [Important new files and their purpose]

**Dependencies:**
- [New packages added]

## Testing
- [X] tests passing
- [Y] tests failing (if any, explain why)

## Next Steps
- [Suggested follow-up actions]

Webdev Checkpoint Checklist

Before creating webdev checkpoint:

  • All tests passing (or failures documented)
  • webdev_check_status shows no critical errors
  • Feature works in dev server
  • todo.md updated with completed items marked [x]
  • Large media files uploaded to S3 (not in project directory)
  • Database migrations applied successfully

Webdev-Specific Guidance

Database Changes

Always follow this sequence:

  1. Edit drizzle/schema.ts
  2. Run pnpm db:push to apply migration
  3. Verify migration succeeded
  4. Test database operations

File Organization

Create files in this order:

  1. Database layer: drizzle/schema.ts
  2. Data access: server/db.ts (query helpers)
  3. Business logic: server/*-router.ts (tRPC endpoints)
  4. Frontend: client/src/components/*.tsx, client/src/pages/*.tsx

Testing Pattern

// server/feature.test.ts
import { describe, it, expect } from 'vitest';

describe('Feature Tests', () => {
  it('should handle valid input', async () => {
    // Arrange
    const input = { /* test data */ };
    
    // Act
    const result = await endpoint(input);
    
    // Assert
    expect(result).toMatchObject({ expected });
  });
  
  it('should handle error cases', async () => {
    // Test error scenarios
  });
});

Common Patterns

Adding a Feature:

  1. Plan → Update todo.md
  2. Schema → Database migration
  3. Backend → tRPC endpoint + tests
  4. Frontend → Component + integration
  5. Test → Manual verification
  6. Checkpoint → Comprehensive description

Fixing a Bug:

  1. Reproduce → Document steps
  2. Identify → Root cause analysis
  3. Fix → Minimal change
  4. Test → Verify fix + no regressions
  5. Checkpoint → Explain bug and fix

Best Practices

Do

✅ Write plans before coding - Saves time, reduces rework
✅ Update todo.md actively - Provides clear progress tracking
✅ Test incrementally - Catch issues early
✅ Document decisions - Future you will thank present you
✅ Create detailed checkpoints - Enable easy rollback and review

Don't

❌ Skip planning - Leads to disorganized implementation
❌ Implement everything at once - Hard to debug, test, and review
❌ Forget to update todo.md - Loses track of progress
❌ Skip testing - Bugs slip into production
❌ Create vague checkpoints - Useless for rollback or review

Templates and Scripts

Available Resources

Templates (templates/):

  • implementation-plan-template.md - Structured feature plan
  • todo-template.md - Format for tracking features

Scripts (scripts/):

  • verify_completion.py - Validates all checklist items complete

References (references/):

  • testing-strategies.md - Testing approaches by project type
  • checkpoint-guidelines.md - Checkpoint best practices

Using Templates

Copy templates to your project:

# Implementation plan
cp /home/ubuntu/skills/systematic-feature-builder/templates/implementation-plan-template.md \
   <project>/docs/plans/$(date +%Y-%m-%d)-feature-name.md

# Todo file (if project doesn't have one)
cp /home/ubuntu/skills/systematic-feature-builder/templates/todo-template.md \
   <project>/todo.md

Examples

Example: Adding Bulk CSV Import Feature

Phase 1: Requirements

  • Admin needs to import rep codes from CSV
  • CSV format: email, repCode
  • Show import results with errors

Phase 2: Plan

### Task 1: Backend Endpoint (30 min)
1. Add bulkImportRepCodes to admin-router.ts
2. Parse CSV data from input
3. Validate each row
4. Update database
5. Return results with errors

### Task 2: Frontend Component (45 min)
1. Create BulkRepCodeImport.tsx
2. Add file upload input
3. Integrate Papa Parse
4. Show import results
5. Add CSV template download

### Task 3: Integration (15 min)
1. Add component to Admin page
2. Test with sample CSV
3. Verify error handling

Phase 3: Implementation

  • Implement tasks sequentially
  • Update todo.md after each task
  • Test each component independently

Phase 4: Testing

  • Unit test: endpoint validates CSV correctly
  • Integration test: upload CSV via UI
  • Manual test: download template, upload, verify results

Phase 5: Checkpoint

Title: Add bulk rep code CSV import

## Features Added
- Admin endpoint for bulk importing rep codes
- BulkRepCodeImport component with file upload
- Papa Parse integration
- Detailed error reporting

## Technical Details
**New Files:**
- server/admin-router.ts: bulkImportRepCodes endpoint
- client/src/components/BulkRepCodeImport.tsx

**Dependencies:**
- papaparse, @types/papaparse

## Testing
- 52 tests passing

Troubleshooting

"Plan is too detailed/taking too long" → Reduce granularity. Focus on major steps, not every line of code.

"Implementation deviating from plan" → Update plan document with deviations and reasoning. Adjust todo.md.

"Tests failing after implementation" → Fix tests before checkpoint. Document any intentional test changes.

"Checkpoint description too long" → Focus on what changed and why. Move detailed technical notes to plan document.

Summary

This skill provides a systematic, organized approach to feature implementation:

  1. Analyze requirements thoroughly
  2. Plan implementation with detailed steps
  3. Implement systematically with progress tracking
  4. Test at multiple levels
  5. Document comprehensively in checkpoint

The result: high-quality, well-tested features with clear documentation and easy rollback capability.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.