work-decomposition
npx machina-cli add skill a5c-ai/babysitter/work-decomposition --openclawWork Decomposition
Overview
Break high-level goals into MEOWs (Molecular Expressions of Work) - the fundamental atomic units in Gas Town. Each MEOW becomes a bead (git-backed work unit) or wisp (ephemeral task).
When to Use
- Before creating a convoy
- When a goal is too large for a single agent
- When parallel execution would benefit progress
- When work needs tracked attribution
Process
- Analyze the goal and project context
- Identify natural seams for decomposition
- Create MEOWs with clear boundaries and dependencies
- Classify as beads (persistent) or wisps (ephemeral)
- Map dependencies between MEOWs
- Estimate effort and assign priorities
Decomposition Principles
- Each MEOW should be completable by a single agent
- Dependencies should form a DAG (no cycles)
- Prefer more smaller beads over fewer larger ones
- Wisps for throwaway work (scaffolding, exploration)
- Every MEOW gets attribution tracking
Tool Use
Invoke via babysitter process: methodologies/gastown/gastown-orchestrator (analyze-work step)
Source
git clone https://github.com/a5c-ai/babysitter/blob/main/plugins/babysitter/skills/babysit/process/methodologies/gastown/skills/work-decomposition/SKILL.mdView on GitHub Overview
Break high-level goals into MEOWs (Molecular Expressions of Work), the atomic units in Gas Town. Each MEOW becomes a bead (git-backed work unit) or a wisp (ephemeral task), enabling clear boundaries, attribution, and parallel execution. This approach keeps work organized, traceable, and scalable across agents.
How This Skill Works
Analyze the goal and project context, identify natural seams for decomposition, and create MEOWs with clear boundaries and dependencies. Classify each MEOW as a bead (persistent) or a wisp (ephemeral), map dependencies to form a DAG, and estimate effort to set priorities. Beads are git-backed units and wisps are throwaway work; every MEOW supports attribution tracking.
When to Use It
- Before creating a convoy
- When a goal is too large for a single agent
- When parallel execution would benefit progress
- When work needs tracked attribution
- When dependencies and boundaries must be explicit
Quick Start
- Step 1: Analyze the goal and project context
- Step 2: Identify natural seams and create MEOWs with boundaries and dependencies
- Step 3: Classify as beads or wisps, map dependencies, estimate effort, and set priorities
Best Practices
- Ensure each MEOW is completable by a single agent
- Design dependencies to form a DAG with no cycles
- Prefer many smaller beads over a few large ones
- Use wisps for scaffolding, exploration, or throwaway tasks
- Track attribution for every MEOW
Example Use Cases
- Decompose a software feature into MEOWs so multiple developers can work in parallel without conflicts
- Break a marketing campaign into BEADS for persistent tasks and WISPs for experimental tasks
- Structure a research project with core beads for foundational work and wisps for exploratory experiments
- Plan a product upgrade convoy with clear MEOW boundaries and dependency mapping
- Manage a refactor by creating a DAG of MEOWs to avoid cycles and ensure attribution