Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

A5c AI/Work Decomposition — measured AI agent skill | DecimalAI

a5c-ai/work-decomposition

SkillScore64Maintained

Decompose goals into MEOWs (Molecular Expressions of Work) - trackable atomic units following Gas Town's bead-based work model.

by a5c-ai

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Project Management pack

decimalai skills pull a5c-ai-work-decomposition

Get this skill — curl · raw file

curl -fsSL https://app.decimal.ai/s/a5c-ai-work-decomposition/SKILL.md

Raw file: /s/a5c-ai-work-decomposition/SKILL.md ↗ · machine-readable summary at /s/a5c-ai-work-decomposition.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 9d ago

Yes — +59% lift-33% tokens0% turns64% of whole cases pass with skill (14/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-01	✗→✓	▲ Improved	-10%	0%
case-02	✗→✓	▲ Improved	7%	0%
case-07	✗→✓	▲ Improved	-57%	0%
case-08	✗→✓	▲ Improved	-35%	0%
case-09	✗→✓	▲ Improved	-66%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 7/31/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

a5c-ai / babysitter

library/methodologies/gastown/skills/work-decomposition/SKILL.md

View on GitHub ↗

License MITImported 7/31/2026Version v1

Skill contents

SKILL.md

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)

Want to see if a5c-ai/work-decomposition would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.