agentleFS
SaehwanParkSaehwanPark/agent-skills-archive/skills/simple-code-writer/SKILL.md

simple-code-writer

Skill1 starsChanged 5 months agoMITView on GitHub

Use when implementing, editing, or refactoring code and the solution should be the smallest correct change, prefer standard or platform capabilities, avoid speculative abstractions, and preserve existing contracts with proportional verification.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 6 are distinctive, and 1 is 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.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ask before adopting a simpler alternative that changes agreed scope or behavior.

    line 26
  2. Distinctive:

    Follow sound local patterns.

    line 32
  3. Distinctive:

    Keep state, effects, ownership, error handling, and data movement visible where they matter.

    line 35
  4. Distinctive:

    Add a dependency only when simpler available options are inadequate and its lifecycle, security, and operational costs are justified.

    line 37
  5. Distinctive:

    Avoid stubs, speculative extension points, drive-by cleanup, and comments that narrate mechanics.

    line 40
  6. Distinctive:

    Run the narrowest relevant tests, type checks, linters, or builds.

    line 51
  7. Very specific:

    Stop when the simpler solution changes a public contract, multiple incompatible architectures exist, a new dependency/migration/security …

    line 63

≈ “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

Simple Code Writer

Choose the lowest total-cost solution across correctness, readability, maintenance, performance, operations, and future change—not merely the fewest lines.

Before editing

Inspect the request, repository instructions, current changes, relevant source and callers, public contracts, tests, and established conventions. Read the simplicity checklist before substantial implementation or refactoring.

Apply these questions in order:

  1. Can the requirement be eliminated, deferred, configured, or enforced at an existing boundary without changing agreed behavior?
  2. Can the standard library or platform solve it directly?
  3. Does an existing dependency solve a problem that belongs to its domain?
  4. What is the clearest direct implementation that satisfies only the current requirement?

Ask before adopting a simpler alternative that changes agreed scope or behavior.

Implementation rules

  • Make the smallest coherent diff that satisfies observable behavior.
  • Preserve public interfaces and user-visible behavior unless change is requested.
  • Follow sound local patterns; do not introduce a competing architecture for a local edit.
  • Introduce a helper or abstraction only when it names a stable concept, removes meaningful duplication, or enforces a real boundary.
  • Keep state, effects, ownership, error handling, and data movement visible where they matter.
  • Add a dependency only when simpler available options are inadequate and its lifecycle, security, and operational costs are justified.
  • Optimize measured hot paths or explicit constraints, not hypothetical ones.
  • Avoid stubs, speculative extension points, drive-by cleanup, and comments that narrate mechanics.

Refactoring

Establish the behavior contract before changing structure. Add characterization coverage when important behavior is unprotected. Prefer local simplification and stop when further work would change behavior, widen scope materially, or require an architectural decision.

Verification

Run the narrowest relevant tests, type checks, linters, or builds. Add focused tests for changed behavior and important unprotected contracts, not implementation details. Report unrelated failures separately.

Composition and conflicts

Narrower domain skills may impose stronger requirements, but every added layer, type, dependency, or process must solve a concrete problem. Follow explicit user and repository instructions when they conflict with this baseline.

Stop conditions

Stop when the simpler solution changes a public contract, multiple incompatible architectures exist, a new dependency/migration/security change is unexpectedly required, or relevant behavior cannot be verified.

Expected output

Report behavior implemented, simpler alternatives materially considered, files changed, tests and checks, deviations, and remaining risks or deferred work.

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.