doc-tasks（文档任务）

doc-tasks

Purpose

Create Task Breakdown (TASKS) - Layer 11 artifact in the SDD workflow that decomposes SPEC into actionable, AI-structured TODO tasks for implementation.

Layer: 11

Upstream: BRD (Layer 1), PRD (Layer 2), EARS (Layer 3), BDD (Layer 4), ADR (Layer 5), SYS (Layer 6), REQ (Layer 7), IMPL (Layer 8), CTR (Layer 9), SPEC (Layer 10)

Downstream Artifacts: Code (Layer 12)

Prerequisites

Upstream Artifact Verification (CRITICAL)

Before creating this document, you MUST:

1. List existing upstream artifacts: ls docs/01_BRD/ docs/02_PRD/ docs/03_EARS/ docs/04_BDD/ docs/05_ADR/ docs/06_SYS/ docs/07_REQ/ 2>/dev/null
2. Reference only existing documents in traceability tags
3. Use null only when upstream artifact type genuinely doesn't exist
4. NEVER use placeholders like BRD-XXX or TBD
5. Do NOT create missing upstream artifacts - skip functionality instead

Before creating TASKS, read:

1. Shared Standards: .claude/skills/doc-flow/SHARED_CONTENT.md
2. Upstream SPEC: Read technical specifications (PRIMARY SOURCE)
3. Template: ai_dev_flow/11_TASKS/TASKS-TEMPLATE.md
4. Creation Rules: ai_dev_flow/11_TASKS/TASKS_CREATION_RULES.md
5. Validation Rules: ai_dev_flow/11_TASKS/TASKS_VALIDATION_RULES.md
6. Validation Script: ./ai_dev_flow/scripts/validate_tasks.sh
7. Implementation Contracts Guide: ai_dev_flow/11_TASKS/IMPLEMENTATION_CONTRACTS_GUIDE.md

When to Use This Skill

Use doc-tasks when:

• Have completed BRD through SPEC (Layers 1-10)
• Ready to break down SPEC into actionable tasks
• Preparing for implementation planning (Layer 12)
• Need structured TODO format for AI agents
• You are at Layer 11 of the SDD workflow

Reserved ID Exemption (TASKS-00_*)

Scope: Documents with reserved ID 000 are FULLY EXEMPT from validation.

Pattern: TASKS-00_*.md

Document Types:

• Index documents (TASKS-00_index.md)
• Traceability matrix templates (TASKS-00_TRACEABILITY_MATRIX-TEMPLATE.md)
• Implementation contracts checklists (TASKS-00_IMPLEMENTATION_CONTRACTS_CHECKLIST.md)
• Glossaries, registries

Rationale: Reserved ID 000 documents are framework infrastructure (indexes, templates, reference materials), not project artifacts requiring traceability or quality gates.

Validation Behavior: Skip all checks when filename matches TASKS-00_* pattern.

Element ID Format (MANDATORY)

Pattern: TASKS.{DOC_NUM}.{HASH} (3 segments, dot-separated)

REMOVED PATTERNS - Do NOT use: - TASK-XXX → Use TASKS.NN.18.SS - T-XXX → Use TASKS.NN.18.SS Reference: ID_NAMING_STANDARDS.md — Cross-Reference Link Format

Fix: Replace ### TASK-01: Implementation with ### TASKS.02.1801: Implementation

TASKS-Specific Guidance

1. AI-Structured TODO Format

Purpose: Break SPEC into tasks consumable by AI coding agents

Format:

## Tasks

### Phase 1: Project Setup (3 tasks)

**TASKS.01.1801: Initialize Project Structure**
- **Action**: Create directory structure per SPEC architecture
- **Files to Create**:
  - `src/controllers/data_validation_controller.py`
  - `src/services/data_validator.py`
  - `src/repositories/data_repository.py`
  - `src/models/data_request.py`
- **Dependencies**: None
- **Estimated Effort**: 30 minutes
- **SPEC Reference**: SPEC-01:implementation.modules
- **Success Criteria**: All directories and empty files created

**TASKS.01.1802: Set Up Development Environment**
- **Action**: Configure Python environment and dependencies
- **Files to Create**: `requirements.txt`, `pyproject.toml`
- **Dependencies**: TASKS.01.1801
- **Estimated Effort**: 1 hour
- **SPEC Reference**: SPEC-01:deployment.container
- **Success Criteria**: `pip install -r requirements.txt` succeeds

### Phase 2: Data Models (2 tasks)

**TASKS.01.1803: Implement DataRequest Model**
- **Action**: Create Pydantic model per CTR-01 schema
- **Files to Modify**: `src/models/data_request.py`
- **Dependencies**: TASKS.01.1802
- **Estimated Effort**: 1 hour
- **SPEC Reference**: SPEC-01:interfaces.data_models
- **CTR Reference**: CTR-01#/components/schemas/DataRequest
- **Success Criteria**: Model validates per schema, unit tests pass

2. Required Sections

Document Control (MANDATORY - First section before all numbered sections)

Core Sections:

1. Overview: Summary of task breakdown
2. Task Hierarchy: Phases and task groups
3. Tasks: Detailed task breakdown (primary content)
4. Dependencies Graph: Visual task dependencies (Mermaid diagram)
5. Effort Summary: Total effort by phase
6. Traceability: Section 7 format with cumulative tags
7. Implementation Contracts: Section 8 (MANDATORY) - Contracts provided/consumed

3. Task Numbering Format

Format: TASKS.{SPEC-ID}.18.{SEQ} (unified element ID format)

Example: TASKS.01.1803 means:

• SPEC-01 (from SPEC-01_data_validation.yaml)
• Element type 18 (Task)
• Sequence 03 (third task in breakdown)

Benefits:

• Links task directly to SPEC
• Unique task IDs across project
• Easy to reference in commits

4. Task Fields (Required)

Each task MUST include:

1. Task ID: TASKS.{SPEC-ID}.18.{SEQ}
2. Title: Short description (5-10 words)
3. Action: What to do (imperative form)
4. Files to Create/Modify: Specific file paths
5. Dependencies: Other TASK IDs (or "None")
6. Estimated Effort: Time estimate
7. SPEC Reference: Section in SPEC (e.g., SPEC-01:implementation.modules)
8. Success Criteria: How to verify completion
9. Optional: CTR Reference: Link to contract if applicable

5. Phase Organization

Typical Phases:

1. Phase 1: Project Setup (infrastructure, environment)
2. Phase 2: Data Models (schemas, models, validation)
3. Phase 3: Business Logic (services, core algorithms)
4. Phase 4: API Layer (controllers, endpoints)
5. Phase 5: Error Handling (error codes, middleware)
6. Phase 6: Configuration (env vars, feature flags)
7. Phase 7: Testing (unit, integration, performance)
8. Phase 8: Deployment (Docker, CI/CD, monitoring)

6. Dependencies Graph

Use Mermaid diagram ONLY (text-based diagrams prohibited per ai_dev_ssd_flow/DIAGRAM_STANDARDS.md):

## Dependencies Graph

graph TD T001[TASKS.01.1801: Project Setup] T002[TASKS.01.1802: Dev Environment] T003[TASKS.01.1803: DataRequest Model] T004[TASKS.01.1804: ValidationResponse Model] T005[TASKS.01.1805: Data Repository] T006[TASKS.01.1806: Data Validator Service] T007[TASKS.01.1807: API Controller]

T001 --> T002 T002 --> T003 T002 --> T004 T002 --> T005 T003 --> T006 T004 --> T006 T005 --> T006 T006 --> T007

7. Effort Summary

Format:

## Effort Summary

| Phase | Tasks | Total Effort |
|-------|-------|--------------|
| Phase 1: Project Setup | 2 | 1.5 hours |
| Phase 2: Data Models | 2 | 2 hours |
| Phase 3: Business Logic | 3 | 4 hours |
| Phase 4: API Layer | 1 | 1.5 hours |
| Phase 5: Error Handling | 2 | 2 hours |
| Phase 6: Configuration | 1 | 1 hour |
| Phase 7: Testing | 3 | 3 hours |
| Phase 8: Deployment | 2 | 2 hours |
| **TOTAL** | **16** | **17 hours** |

**Assumptions**:
- Developer familiar with Python and FastAPI
- PostgreSQL database already provisioned
- OAuth service already available

8. Implementation Contracts (MANDATORY)

Section 8 is required for ALL TASKS files. Implementation Contracts enable parallel development of dependent TASKS files.

Structure:

## 8. Implementation Contracts

### 8.1 Contracts Provided by This TASKS
@icon: TASKS-XXX:ContractName
@icon-role: provider

- **Contract Name**: [Interface name]
- **Type**: Protocol Interface | Exception Hierarchy | State Machine | Data Model | DI Interface
- **Consumers**: List of TASKS IDs that depend on this contract
- **Purpose**: Brief description

### 8.2 Contracts Consumed by This TASKS
@icon: TASKS-YYY:OtherContract
@icon-role: consumer

- **Provider**: TASKS-YYY
- **Contract Name**: [Interface name]
- **Purpose**: Why this TASKS needs this contract

### 8.3 No Contracts
If this TASKS provides no contracts and consumes no contracts, state explicitly:
"This TASKS document neither provides nor consumes implementation contracts."

When to Create Contracts:

• TASKS has 3+ downstream dependencies
• Shared interfaces across multiple implementation sessions
• Complex state machines or exception hierarchies
• Parallel development required

Contract Types:

1. Protocol Interfaces: typing.Protocol with method signatures
2. Exception Hierarchies: Typed exceptions with error codes
3. State Machine Contracts: Enum states with valid transitions
4. Data Models: Pydantic/TypedDict schemas
5. DI Interfaces: ABC classes for dependency injection

Reference: See ai_dev_flow/11_TASKS/IMPLEMENTATION_CONTRACTS_GUIDE.md for detailed guidance.

Tag Format Convention (By Design)

The SDD framework uses two distinct notation systems for cross-references:

Key Distinction:

• @adr: ADR-033 → Points to the document ADR-033_risk_limit_enforcement.md
• @brd: BRD.17.0101 → Points to element 01.01 inside document BRD-017.md

Unified Element ID Format (MANDATORY)

For hierarchical requirements (BRD, PRD, EARS, BDD, SYS, REQ):

• Always use: TYPE.NN.xxxx (dot separator, 3-segment unified format)
• Never use: TYPE-NN:NNN (colon separator - DEPRECATED)
• Never use: TYPE.NN.TT.SS (old 4-segment format - DEPRECATED)

Examples:

• @brd: BRD.17.0101 ✅
• @brd: BRD.017.001 ❌ (old 3-segment format)

Cumulative Tagging Requirements

Layer 11 (TASKS): Must include tags from Layers 1-10

Tag Count: 8-10 tags (minimum 8, maximum 10)

Element Type Codes for Cumulative Tags

Minimum (IMPL and CTR skipped):

## Traceability

**Required Tags** (Cumulative Tagging Hierarchy - Layer 11):

@brd: BRD.01.0103 @prd: PRD.01.0702 @ears: EARS.01.2501 @bdd: BDD.01.1401 @adr: ADR-033, ADR-045 @sys: SYS.01.2601 @req: REQ.01.2701 @spec: SPEC-01

**Maximum (IMPL, CTR, and ICON included)**:

@brd: BRD.01.0103 @prd: PRD.01.0702 @ears: EARS.01.2501 @bdd: BDD.01.1401 @adr: ADR-033, ADR-045 @sys: SYS.01.2601 @req: REQ.01.2701 @impl: IMPL.01.2901 @ctr: CTR-01 @spec: SPEC-01 @icon: TASKS-01:DataValidator # if providing or consuming implementation contracts @icon-role: provider # or consumer

## Upstream/Downstream Artifacts

**Upstream Sources**:

- **BRD** (Layer 1) - Business requirements
- **PRD** (Layer 2) - Product features
- **EARS** (Layer 3) - Formal requirements
- **BDD** (Layer 4) - Test scenarios
- **ADR** (Layer 5) - Architecture decisions
- **SYS** (Layer 6) - System requirements
- **REQ** (Layer 7) - Atomic requirements
- **IMPL** (Layer 8) - Implementation approach (optional)
- **CTR** (Layer 9) - Data contracts (optional)
- **SPEC** (Layer 10) - Technical specifications (PRIMARY SOURCE)

**Downstream Artifacts**:

- **Code** (Layer 12) - Implementation code (created in `src/`)
- **Tests** (Layer 13) - Test suites

**Same-Type Document Relationships** (conditional):

- `@related-tasks: TASKS-NN` - TASKS sharing implementation context
- `@depends-tasks: TASKS-NN` - TASKS that must be completed first

## Validation Checks

### Tier 1: Errors (Blocking)

| Check | Description |
| --- | --- |
| CHECK 1 | Filename format valid (TASKS-NN_slug_tasks.md) |
| CHECK 2 | YAML frontmatter present with required fields |
| CHECK 3 | Document Control table complete (8 fields) |
| CHECK 4 | All 8 required sections present |
| CHECK 5 | Section 8 (Implementation Contracts) exists with 8.1/8.2/8.3 subsection |
| CHECK 6 | All 8 required traceability tags present |
| CHECK 7 | Parent SPEC reference valid and file exists |
| CHECK 8 | Element ID format compliance (TASKS.NN.xxxx) |

### Tier 2: Warnings

| Check | Description |
| --- | --- |
| CHECK W1 | Scope section has exclusions documented |
| CHECK W2 | Plan has at least 3 numbered steps |
| CHECK W3 | SPEC line references present |
| CHECK W4 | At least 3 acceptance criteria checkboxes |
| CHECK W5 | BDD scenario reference in acceptance criteria |
| CHECK W6 | @icon-role tag present if @icon tag used |

### Tier 3: Info

| Check | Description |
| --- | --- |
| CHECK I1 | Time estimates present in plan |
| CHECK I2 | Optional @impl and @ctr tags present |
| CHECK I3 | ICON references valid and files exist |

## Creation Process

### Step 1: Read Upstream SPEC

Read SPEC (Layer 10) - technical specifications to decompose.

### Step 2: Reserve ID Number

Check `docs/11_TASKS/` for next available ID number.

**ID Numbering Convention**: Start with 2 digits and expand only as needed.

- ✅ Correct: TASKS-01, TASKS-99, TASKS-102
- ❌ Incorrect: TASKS-001, TASKS-009 (extra leading zero not required)

**ID Matching**: TASKS ID typically matches SPEC ID (TASKS-01 from SPEC-01).

### Step 3: Create TASKS File

**Nested Folder Rule (MANDATORY)**: ALL TASKS documents MUST use nested folders regardless of document size.

**File naming**: `docs/11_TASKS/TASKS-NN_{slug}/TASKS-NN_{slug}.md`

**Example**: `docs/11_TASKS/TASKS-01_data_validation/TASKS-01_data_validation.md`

**CRITICAL**: Never create TASKS files directly in `docs/11_TASKS/` without a nested folder structure.

### Step 4: Fill Document Control Section

Complete metadata and Document Revision History table.

### Step 5: Write Overview

Summarize task breakdown approach.

### Step 6: Define Phases

Organize tasks into logical phases (8 typical phases).

### Step 7: Create Detailed Tasks

For each task in SPEC:

- Assign TASK ID (TASKS.{SPEC-ID}.18.{SEQ})
- Write clear Action (imperative)
- List Files to Create/Modify
- Identify Dependencies
- Estimate Effort
- Reference SPEC section
- Define Success Criteria

### Step 8: Create Dependencies Graph

Use Mermaid diagram to visualize task dependencies.

### Step 9: Calculate Effort Summary

Summarize total effort by phase.

### Step 10: Add Cumulative Tags

Include all 8-10 upstream tags (@brd through @spec).

### Step 11: Create/Update Traceability Matrix

**MANDATORY**: Update `docs/TASKS/TASKS-00_TRACEABILITY_MATRIX.md`

### Step 12: Validate TASKS

./ai_dev_flow/scripts/validate_tasks.sh docs/TASKS/TASKS-01_*.md

python ai_dev_flow/scripts/validate_tags_against_docs.py --artifact TASKS-01 --expected-layers brd,prd,ears,bdd,adr,sys,req,impl,contracts,spec --strict

### Step 13: Commit Changes

Commit TASKS file and traceability matrix.

## Validation

### Automated Validation

Quality gates

./scripts/validate_quality_gates.sh docs/TASKS/TASKS-01_*.md

Task format validation

./ai_dev_flow/scripts/validate_tasks.sh docs/TASKS/TASKS-01_*.md

Cumulative tagging

python ai_dev_flow/scripts/validate_tags_against_docs.py \ --artifact TASKS-01 \ --expected-layers brd,prd,ears,bdd,adr,sys,req,impl,contracts,spec \ --strict

### Manual Checklist

- Document Control section at top
- Overview explains breakdown approach
- Tasks organized into phases (8 typical phases)
- Each task has TASKS.{SPEC-ID}.18.{SEQ} ID
- Each task has all required fields
- Dependencies identified (or "None")
- Effort estimates provided
- SPEC references included
- Success Criteria clear and testable
- Dependencies Graph (Mermaid diagram) created
- Effort Summary calculated
- **Section 8 Implementation Contracts** completed (provider/consumer/none)
- Cumulative tags: @brd through @spec (8-10 tags) included
- `@icon` tags added if providing/consuming contracts
- Traceability matrix updated

### Diagram Standards

All diagrams MUST use Mermaid syntax. Text-based diagrams (ASCII art, box drawings) are prohibited. See: `ai_dev_ssd_flow/DIAGRAM_STANDARDS.md` and `mermaid-gen` skill.

## Common Pitfalls

1. **Vague tasks**: Tasks must be specific and actionable
2. **Missing dependencies**: Must identify task dependencies
3. **No effort estimates**: Effort required for planning
4. **Missing SPEC references**: Each task must link to SPEC section
5. **No success criteria**: Must define how to verify completion
6. **Missing cumulative tags**: Layer 11 must include all 8-10 upstream tags
7. **Missing Section 8**: Implementation Contracts section is MANDATORY
8. **Text-based diagrams**: Use Mermaid ONLY for Dependencies Graph
9. **Legacy task IDs**: Use TASKS.NN.18.SS, NOT TASK-XXX or T-XXX

## Post-Creation Validation (MANDATORY - NO CONFIRMATION)

**CRITICAL**: Execute this validation loop IMMEDIATELY after document creation. Do NOT proceed to next document until validation passes.

### Automatic Validation Loop

LOOP: 1. Run: python ai_dev_flow/scripts/validate_cross_document.py --document {doc_path} --auto-fix 2. IF errors fixed: GOTO LOOP (re-validate) 3. IF warnings fixed: GOTO LOOP (re-validate) 4. IF unfixable issues: Log for manual review, continue 5. IF clean: Mark VALIDATED, proceed

### Validation Command

Per-document validation (Phase 1)

python ai_dev_flow/scripts/validate_cross_document.py --document docs/TASKS/TASKS-NN_slug.md --auto-fix

Layer validation (Phase 2) - run when all TASKS documents complete

python ai_dev_flow/scripts/validate_cross_document.py --layer TASKS --auto-fix

### Layer-Specific Upstream Requirements

| This Layer | Required Upstream Tags | Count |
| --- | --- | --- |
| TASKS (Layer 11) | @brd, @prd, @ears, @bdd, @adr, @sys, @req, @spec (+ @impl, @ctr if created) | 8-10 tags |

### Auto-Fix Actions (No Confirmation Required)

| Issue | Fix Action |
| --- | --- |
| Missing upstream tag | Add with upstream document reference |
| Invalid tag format | Correct to TYPE.NN.xxxx (3-segment) or TYPE-NN format |
| Broken link | Recalculate path from current location |
| Missing traceability section | Insert from template |

### Validation Codes Reference

| Code | Description | Severity |
| --- | --- | --- |
| XDOC-001 | Referenced requirement ID not found | ERROR |
| XDOC-002 | Missing cumulative tag | ERROR |
| XDOC-003 | Upstream document not found | ERROR |
| XDOC-006 | Tag format invalid | ERROR |
| XDOC-007 | Gap in cumulative tag chain | ERROR |
| XDOC-009 | Missing traceability section | ERROR |

### Quality Gate

**Blocking**: YES - Cannot proceed to next document until Phase 1 validation passes with 0 errors.

---

## Next Skill

After completing TASKS, proceed to implementation:

- **Code** (Layer 12) - Write implementation code
- **Tests** (Layer 13) - Create test suites
- **Validation** (Layer 14) - Run validation checks

## Reference Documents

TASKS artifacts do not support REF documents. Reference documents are limited to **BRD and ADR types only** per the SDD framework.

For supplementary documentation needs, create:

- **BRD-REF**: Business context documentation
- **ADR-REF**: Task estimation guides, dependency analysis reports

## Related Resources

- **Template**: `ai_dev_flow/11_TASKS/TASKS-TEMPLATE.md` (primary authority)
- **TASKS Creation Rules**: `ai_dev_flow/11_TASKS/TASKS_CREATION_RULES.md`
- **TASKS Validation Rules**: `ai_dev_flow/11_TASKS/TASKS_VALIDATION_RULES.md`
- **TASKS README**: `ai_dev_flow/11_TASKS/README.md`
- **Shared Standards**: `.claude/skills/doc-flow/SHARED_CONTENT.md`

**Section Templates** (for documents >25K tokens):

- Index template: `ai_dev_flow/11_TASKS/TASKS-SECTION-0-TEMPLATE.md`
- Content template: `ai_dev_flow/11_TASKS/TASKS-SECTION-TEMPLATE.md`
- Reference: `ai_dev_flow/ID_NAMING_STANDARDS.md` (Section-Based File Splitting)

## Quick Reference

**TASKS Purpose**: Decompose SPEC into actionable AI-structured TODO tasks

**Layer**: 11

**Element ID Format**: `TASKS.NN.18.SS`

- Task = 18
- Task Item = 30

**Removed Patterns**: TASK-XXX, T-XXX

**Tags Required**: @brd through @spec (8-10 tags)

**Format**: AI-structured TODO with phases

**Task ID Format**: TASKS.{SPEC-ID}.18.{SEQ}

**Required Task Fields**:

- Task ID, Title, Action
- Files to Create/Modify
- Dependencies
- Estimated Effort
- SPEC Reference
- Success Criteria

**Key Sections**:

- Task Hierarchy (phases)
- Detailed Tasks (primary content)
- Dependencies Graph (Mermaid)
- Effort Summary
- Implementation Contracts (Section 8 - MANDATORY)

**Next**: Implementation (Code → Tests → Validation)

---

## Version History

| Version | Date | Changes | Author |
| --- | --- | --- | --- |
| 1.0 | 2026-02-08 | Initial skill definition with YAML frontmatter standardization | System |