agentleFS
b1tankb1tank/skills/.github/skills/decompose-task/SKILL.md

decompose-task

Skill1 starsChanged 8 months agoMITView on GitHub

Break large tasks (more than 100 lines) into atomic, verifiable sub-tasks. Use when a task scope is unclear, touches multiple files/modules, or exceeds comfortable single-commit size. Returns structured breakdown without implementing code.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Analyze scope: Read relevant files, estimate lines per area

    line 20
  2. Distinctive:

    Propose sub-tasks: Create checkbox list with line estimates

    line 22
  3. Distinctive:

    Do NOT implement code—only analyze and decompose

    line 53
  4. Distinctive:

    Do NOT update plan files directly—return proposal for human confirmation

    line 54
  5. Distinctive:

    If task is already small (<100 lines), report that no decomposition is needed

    line 55
  6. Distinctive:

    Prefer vertical slices (end-to-end thin features) over horizontal layers

    line 56

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Task Decomposition

Analyze a task and decompose it into atomic, verifiable sub-tasks when scope exceeds ~100 lines.

When to Use

  • Estimated >100 lines of code change
  • Multiple unrelated files changing
  • Multiple state transitions or new states
  • Backend + frontend changes beyond contract sync
  • Vague descriptions ("implement pipeline", "add full support")

Process

  1. Analyze scope: Read relevant files, estimate lines per area
  2. Identify boundaries: Find natural cut points (interfaces, modules, layers)
  3. Propose sub-tasks: Create checkbox list with line estimates
  4. Return proposal for human confirmation

Sub-Task Criteria

Each sub-task should be:

  • Completable in one atomic commit
  • ~10-50 lines (occasionally up to 100)
  • Independently testable or verifiable
  • Clearly scoped with success criteria

Output Format

[DECOMPOSITION COMPLETE]

Original task: [description]
Estimated total scope: ~[N] lines across [M] files

Proposed sub-tasks:
- [ ] [sub-task 1] (~N lines) - [brief rationale]
- [ ] [sub-task 2] (~N lines) - [brief rationale]
- [ ] ...

Recommended order: [1 → 2 → 3] or [1, 2 can parallel → 3]
Dependencies: [any blockers or prerequisites]
Questions for human: [if any]

Guidelines

  • Do NOT implement code—only analyze and decompose
  • Do NOT update plan files directly—return proposal for human confirmation
  • If task is already small (<100 lines), report that no decomposition is needed
  • Prefer vertical slices (end-to-end thin features) over horizontal layers

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.