work-unit-decomposition
npx machina-cli add skill a5c-ai/babysitter/work-unit-decomposition --openclawWork Unit Decomposition
Overview
Break implementation plans into discrete, testable work units. Each unit has enumerated Definition of Done items, declared file scope, dependency mapping, and optional human checkpoint flags.
When to Use
- After plan review gate approval
- When decomposing a large feature into implementable chunks
- When preparing for orchestrated execution
Work Unit Structure
Each work unit specifies:
- ID - Unique identifier
- Title - Human-readable description
- Definition of Done - Enumerated checklist items
- File Scope - Which files the unit may modify
- Dependencies - Other work units this depends on
- Human Checkpoint - Whether human approval is needed before execution
- Parallel Safe - Whether it can execute alongside other units
Human Checkpoint Triggers
Mark human checkpoints for:
- Schema or database changes
- Security-sensitive code paths
- New architectural patterns not seen in codebase
- External API integrations
- Breaking changes to public interfaces
Agents Used
agents/architect/- Creates the decompositionagents/cto/- Validates TDD readiness per unit
Tool Use
Invoke as part of: methodologies/metaswarm/metaswarm-orchestrator (Phase 4)
Source
git clone https://github.com/a5c-ai/babysitter/blob/main/plugins/babysitter/skills/babysit/process/methodologies/metaswarm/skills/work-unit-decomposition/SKILL.mdView on GitHub Overview
Break implementation plans into discrete, testable work units. Each unit includes an enumerated Definition of Done, declared file scope, dependency mapping, and optional human checkpoint flags. This approach enables safer orchestration and measurable progress.
How This Skill Works
Each work unit defines ID, Title, Definition of Done, File Scope, Dependencies, Human Checkpoint, and Parallel Safe. When ready to execute, the metaswarm orchestrator (Phase 4) uses agents/architect/ to create the decomposition and agents/cto/ to validate TDD readiness per unit.
When to Use It
- After plan review gate approval
- When decomposing a large feature into implementable chunks
- When preparing for orchestrated execution
- When aligning dependencies and potential parallelization across units
- When introducing new architectural patterns or external API integrations
Quick Start
- Step 1: Identify the high-level plan and divide into candidate work units
- Step 2: For each unit, specify ID, Title, DoD, File Scope, Dependencies, and Human Checkpoint; set Parallel Safe
- Step 3: Run the metaswarm orchestrator (Phase 4) and validate TDD readiness with agents
Best Practices
- Define a unique ID and a human-readable Title for each unit
- Enumerate an explicit Definition of Done checklist
- Declare exact File Scope and allowed modifications up to that boundary
- Map Dependencies clearly and flag parallel safety
- Flag Human Checkpoints for high-risk changes (security, schema, API, breaking changes)
Example Use Cases
- Decomposing a new authentication feature into units with DoD
- Migrating a database schema with clear DoD per change
- Breaking a microservice feature into independently testable units
- Implementing a new external API integration with a human checkpoint at API contract changes
- Coordinating a UI refactor with explicit file scope declarations