agentleFS
arumaekawaarumaekawa/codex-skills/skills/build-incremental/SKILL.md

build-incremental

Skill4 starsChanged 4 months agoMITView on GitHub

Use during implementation after a concrete design exists to build in practical behavior slices, keeping each step aligned with the design while avoiding overly tiny steps, unused scaffolding, unnecessary helpers, adapters, fallbacks, compatibility layers, and broad input handling.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 11 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈11

    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:

    Use this skill during implementation to build the designed behavior in practical steps.

    line 8
  2. Distinctive:

    Read the concrete implementation design provided by the caller.

    line 12
  3. Distinctive:

    Run the smallest useful verification when practical.

    line 16
  4. Distinctive:

    If the design becomes incomplete or incorrect, revise it or return it for revision before continuing.

    line 17
  5. Distinctive:

    Prefer clear, readable implementation over clever, overly generic, or compact code.

    line 36
  6. Distinctive:

    Add concise comments where they clarify non-obvious intent, constraints, or reasoning.

    line 37
  7. Distinctive:

    Do not route around existing code with exception paths, side helpers, adapters, or compatibility layers.

    line 38
  8. Distinctive:

    Do not split logic into tiny functions before responsibilities become clear.

    line 39
  9. Distinctive:

    Do not add fallback behavior the user did not request.

    line 40
  10. Distinctive:

    Use clear errors for unsupported or unexpected states instead of guessing or silently continuing.

    line 41
  11. Distinctive:

    Do not add unused scaffolding or broaden input handling beyond the design.

    line 42

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Build Incremental

Use this skill during implementation to build the designed behavior in practical steps. Prefer coherent behavior slices over file-based or mechanically tiny steps.

Workflow

  1. Read the concrete implementation design provided by the caller.
  2. Choose the next practical behavior slice.
  3. Implement only that slice while following the implementation guardrails.
  4. Confirm the step still matches the design and implementation guardrails.
  5. Run the smallest useful verification when practical.
  6. If the design becomes incomplete or incorrect, revise it or return it for revision before continuing.
  7. Record or return the changes, verification results, and material discoveries as directed by the caller.
  8. Repeat until the designed behavior is implemented.

Step Size

A step should be a coherent behavior slice that is small enough to review and verify, but not so small that it fragments readable implementation.

Good boundaries include:

  • one normal behavior path
  • one failure path
  • one integration point
  • one local, behavior-preserving structure change that prepares the next behavior slice

Implementation Guardrails

While implementing each step:

  • Prefer clear, readable implementation over clever, overly generic, or compact code.
  • Add concise comments where they clarify non-obvious intent, constraints, or reasoning; avoid comments that merely restate the code.
  • Do not route around existing code with exception paths, side helpers, adapters, or compatibility layers; improve the local structure when needed.
  • Do not split logic into tiny functions before responsibilities become clear.
  • Do not add fallback behavior the user did not request.
  • Use clear errors for unsupported or unexpected states instead of guessing or silently continuing.
  • Do not add unused scaffolding or broaden input handling beyond the design.

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.