tf-task-patterns

name	tf-task-patterns
description	Task breakdown format and phase organization for Terraform implementation.

Task Breakdown Patterns

Transform plan.md into dependency-ordered, checklist-format tasks organized by implementation phase. The tf-implement orchestrator passes phases to tf-task-executor subagents.

Workflow

1. Extract: Pull user stories, modules, data model, infrastructure from plan.md and spec.md
2. Map requirements: Build coverage matrix linking requirements → tasks
3. Assign phases: Setup → Foundational → User Stories (priority order) → Polish
4. Order tasks: Sequential T001, T002... respecting dependencies
5. Label stories: Add [US#] labels only in story phases
6. Generate sections: Header, matrix, phases, dependencies, strategy, checklist, summary

Output

• Location: tasks.md in feature directory
• Format: Markdown checklist with phase headers

Required Sections

Section	Purpose
Header	Feature name, input path, prerequisites, tests stance, organization note
Format explanation	[ID] [Story] Description with sequential execution note
Requirements Coverage Matrix	Requirement → Task(s) → Description traceability
Phase sections	Grouped tasks with purpose, checkpoints, and dependency notes
Dependencies & Execution Order	Phase deps, story deps, cross-module data flow table
Implementation Strategy	MVP first, incremental delivery approach
File Checklist	File → Task → Purpose mapping
Task Summary	Phase → Task range → User Story table with total count

Constraints

• Every task: - [ ] T### [US#?] Description with file path
• Include checkpoint markers after each phase
• Document circular dependencies explicitly

Examples

Good task format:

- [ ] T001 Create project structure per implementation plan
- [ ] T005 [US1] Implement auth middleware in src/middleware/auth.py
- [ ] T010 [US1] Implement CloudFront module in main.tf with OAI creation, S3 origin at /main.tf

Good phase header:

## Phase 2: Foundational (Blocking Prerequisites)

**Purpose**: Core S3 bucket infrastructure that MUST be complete before CloudFront

**CRITICAL**: CloudFront origin configuration requires S3 bucket outputs

- [ ] T005 Implement S3 bucket module in main.tf with public access blocking at /main.tf

**Checkpoint**: S3 bucket module configured - CloudFront implementation can proceed

Bad:

- [ ] Create User model

Missing task ID, story label, and file path.

Context

Phase Structure

Phase	Content	Story Labels
1 - Setup	Project initialization, file structure, Terraform config	No
2 - Foundational	Blocking prerequisites, core infrastructure	No
3+ - User Stories	Priority order (P1, P2...) with independent tests	Required
Final - Polish	Cross-cutting concerns, formatting, validation	No

Source Material Placement

Source	Placement
User stories	Own phase; map models, services, endpoints, tests
Module contracts	Serving story; interface tests before impl if TDD
Data model	Earliest needing story; multi-story entities → Setup
Infrastructure	Shared → Phase 1; blocking → Phase 2; story-specific → that phase

Header Template

# Tasks: [Feature Name]

**Input**: [spec path]
**Prerequisites**: plan.md, spec.md, data-model.md, contracts/module-interfaces.md
**Tests**: [from spec or "No tests requested"]
**Organization**: [brief description]

Cross-Module Data Flow Table

| Task | From | Output | To | Input |
|------|------|--------|-----|-------|
| T010 | `module.s3_bucket` | `bucket_regional_domain_name` | `module.cloudfront` | `origin.domain_name` |

Task Summary Table

| Phase | Tasks | User Story |
|-------|-------|------------|
| Phase 1: Setup | T001-T004 | - |
| Phase 2: Foundational | T005-T007 | - |
| Phase 3: User Story 1 | T008-T013 | US1 (P1) |

**Total Tasks**: N