Skip to content
Back to skills

Create Story Tasks

ASecurity

Use to decompose user stories into individual development tasks. Creates task documents that can be assigned, estimated, and tracked.

  • 666 stars
  • 0 votes
  • 1 copy
  • 5 views
  • Added February 8, 2026
developmentgotestingapidatabasefrontendbackendsecurityperformancedocumentation

Works with

  • api

Security analysis

A100/100

Scanned February 12, 2026

npx -y skills add majiayu000/claude-skill-registry --skill create-story-tasks --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Create Story Tasks?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Create Story Tasks
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/majiayu000-create-story-tasks/badge)](https://www.skillsdirectory.com/skills/majiayu000-create-story-tasks)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: create-story-tasks
description: Use to decompose user stories into individual development tasks. Creates task documents that can be assigned, estimated, and tracked.
version: 1.0.0
---
<!-- Powered by Prism Coreâ„¢ -->

# create-story-tasks

Break down a user story into individual development tasks, creating task documents for each.

## When to Use

- After user story is drafted and approved
- When preparing work breakdown for sprint planning
- When story needs granular task estimation
- When assigning work to multiple developers
- When tracking progress at task level

## Quick Start

1. Load user story from docs/stories/
2. Analyze story scope and acceptance criteria
3. Map ACs to implementation work and dependencies
4. Generate task breakdown (2-8 hour tasks)
5. Create task documents with estimates

## Purpose

Decompose a user story into specific, actionable development tasks that can be assigned, estimated, and tracked independently. Each task should be small enough to complete in a single development session but meaningful enough to deliver value toward the story's acceptance criteria.

## SEQUENTIAL Task Execution

### 1. Load User Story
```yaml
inputs:
  - Story document from docs/stories/
  - Story ID, title, and description
  - Acceptance criteria list
  - Technical notes and requirements
  - Initial task breakdown from story
```

### 2. Analyze Story Scope

```yaml
analysis:
  acceptance_criteria_coverage:
    - Map each AC to required implementation work
    - Identify technical components affected
    - Determine task dependencies
    
  technical_assessment:
    - Frontend components needed
    - Backend services/APIs needed
    - Database changes required
    - Testing requirements
    
  task_granularity:
    - Each task should be 2-8 hours of work
    - Tasks should be independently testable
    - Tasks should align with PRISM principles
```

### 3. Generate Task Breakdown

```yaml
task_structure:
  naming_convention: "{{story_id}}-T{{task_num}}"
  # Example: US-001-T1, US-001-T2, etc.
  
  task_categories:
    - Database/Schema changes
    - Backend API implementation
    - Frontend component development
    - Testing implementation
    - Documentation updates
    - Integration/deployment tasks
    
  task_template_per_task:
    task_id: "{{story_id}}-T{{number}}"
    task_title: "{{brief_description}}"
    acceptance_criteria_refs: [{{ac_numbers}}]
    estimated_hours: "{{2-8}}"
    dependencies: [{{other_task_ids}}]
```

### 4. Create Individual Task Documents

For each identified task, create a task document:

```markdown
# Task {{story_id}}-T{{number}}: {{task_title}}

## Story Reference
- **Parent Story:** {{story_id}} - {{story_title}}
- **Story Link:** [{{story_id}}](../stories/{{story_file}})

## Task Description

{{detailed_description_of_what_needs_to_be_done}}

## Acceptance Criteria Coverage

This task contributes to the following story acceptance criteria:
- AC #{{number}}: {{criteria_text}}
- AC #{{number}}: {{criteria_text}}

## Implementation Requirements

### Functional Requirements
1. {{requirement_1}}
2. {{requirement_2}}
3. {{requirement_3}}

### Technical Details
- **Component/Module:** {{component_name}}
- **Files to Modify:** 
  - {{file_path_1}}
  - {{file_path_2}}
- **New Files to Create:**
  - {{file_path_3}}

### PRISM Principles Application
- **Predictability:** {{how_task_ensures_predictable_outcome}}
- **Resiliency:** {{testing_strategy}}
- **Intentionality:** {{design_decisions_and_rationale}}
- **Sustainability:** {{maintainable_approach}}
- **Maintainability:** {{clear_code_structure}}

## Dependencies

### Upstream Dependencies (must complete first)
- {{task_id}}: {{brief_reason}}

### Downstream Dependencies (waiting on this)
- {{task_id}}: {{brief_reason}}

### External Dependencies
- {{service/resource}}: {{what_is_needed}}

## Testing Requirements

### Unit Tests
- Test: {{test_case_1}}
- Test: {{test_case_2}}

### Integration Tests
- Test: {{integration_scenario}}

### Manual Testing Steps
1. {{step_1}}
2. {{step_2}}
3. {{step_3}}

## Definition of Done
- [ ] Implementation complete per requirements
- [ ] Unit tests written and passing
- [ ] Integration tests passing
- [ ] Code reviewed and approved
- [ ] Documentation updated
- [ ] Acceptance criteria validated
- [ ] No regression in existing functionality

## Estimation

### PSP/PROBE Estimation
- **Estimated Hours:** {{optimistic}}-{{likely}}-{{pessimistic}}
- **Story Points:** {{1|2|3}} (task level)
- **Size Category:** {{very_small|small|medium}}
- **Similar Tasks:** {{references}}
- **Confidence:** {{high|medium|low}}

### Tracking
- **Started:** {{date_time}}
- **Completed:** {{date_time}}
- **Actual Hours:** {{hours}}
- **Variance:** {{percentage}}

## Technical Notes

{{any_architecture_patterns_to_follow}}
{{coding_standards_references}}
{{security_considerations}}
{{performance_requirements}}

## Implementation Notes

{{dev_agent_records_implementation_details}}
{{challenges_encountered}}
{{decisions_made}}
{{files_actually_modified}}

---

**Status:** {{Not Started|In Progress|Review|Testing|Done}}
**Assigned To:** {{agent_or_developer}}
**Created:** {{date}}
**Updated:** {{date}}
```

### 5. Task Ordering Strategy

```yaml
recommended_task_order:
  phase_1_foundation:
    - Database schema changes
    - Data models/entities
    - Migration scripts
    
  phase_2_backend:
    - API endpoints
    - Business logic
    - Service layer
    
  phase_3_frontend:
    - UI components
    - State management
    - Integration with APIs
    
  phase_4_testing:
    - E2E tests
    - Integration tests
    - Performance tests
    
  phase_5_finalization:
    - Documentation
    - Deployment configuration
    - Monitoring/observability

dependency_rules:
  - Backend tasks before frontend tasks
  - Database changes before API changes
  - Component creation before integration
  - Implementation before testing
  - Testing before documentation
```

### 6. Create Task Index

Update the parent story document:

```markdown
## Tasks Created

| Task ID | Title | Status | Estimated | Actual | AC Refs |
|---------|-------|--------|-----------|--------|---------|
| {{story_id}}-T1 | {{title}} | {{status}} | {{est_hours}} | {{actual_hours}} | #1, #2 |
| {{story_id}}-T2 | {{title}} | {{status}} | {{est_hours}} | {{actual_hours}} | #3 |
| {{story_id}}-T3 | {{title}} | {{status}} | {{est_hours}} | {{actual_hours}} | #2, #4 |

**Total Estimated Hours:** {{sum_of_estimates}}
**Total Actual Hours:** {{sum_of_actuals}}
**Task Count:** {{number_of_tasks}}
```

### 7. Quality Checks

```yaml
validation_checklist:
  coverage:
    - [ ] All acceptance criteria mapped to tasks
    - [ ] All technical components addressed
    - [ ] Testing tasks included
    - [ ] Documentation tasks present
    
  task_quality:
    - [ ] Each task has clear, actionable description
    - [ ] Dependencies identified and documented
    - [ ] Estimates provided using PROBE method
    - [ ] Testing requirements specified
    - [ ] PRISM principles applied
    
  sizing:
    - [ ] No task exceeds 8 hours estimate
    - [ ] Tasks are independently deliverable
    - [ ] Task breakdown is logical and sequential
```

## Success Criteria
- [ ] All story acceptance criteria covered by tasks
- [ ] Task documents created in docs/tasks/
- [ ] Task dependencies mapped
- [ ] Estimates provided for all tasks
- [ ] Parent story document updated with task index
- [ ] Tasks ordered by logical implementation sequence
- [ ] Each task aligns with PRISM principles

## Output
- Task document files: `docs/tasks/{{story_id}}-T{{n}}-{{slug}}.md`
- Updated story document with task index
- Task dependency graph (if complex)
- Total effort estimate for story

## Integration with PRISM Workflow

This task integrates with:
- **Before:** Story creation and refinement
- **After:** Individual task execution by Dev agent
- **Parallel:** QA test design from acceptance criteria
- **Quality Gates:** Story estimation, sprint planning

## Notes

- Keep tasks focused and atomic
- Ensure clear handoff between tasks
- Document assumptions and constraints
- Use consistent naming conventions
- Reference architecture and coding standards
- Consider rollback/revert strategy for each task

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.