agentleFS
kadaliaokadaliao/skills/minimalist-coding/SKILL.md

minimalist-coding

Skill0 starsChanged 4 months agoMITView on GitHub

Focused implementation workflow that chooses the smallest clear, compatible, and risk-proportionate change after checking whether the requested capability already exists. Trigger only when the user explicitly invokes /minimalist-coding or asks for minimalist coding by name; never auto-run for ordinary implementation, bug-fix, simplification, or refactoring requests.

The file

Show its rules and text

We split this skill into the rules it gives an agent

17 rules. About 0 are common, 17 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈17

    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:

    Always respond in the user's language.

    line 16
  2. Distinctive:

    Before adding code, inspect enough of the local code and constraints to choose the first option that satisfies the request:

    line 20
  3. Distinctive:

    Use the standard library: - prefer it when it solves the problem clearly.

    line 25
  4. Distinctive:

    Use an installed dependency: - do not add a package when an existing dependency fits.

    line 27
  5. Distinctive:

    Add minimal code: - write only the code needed to meet the defined behavior.

    line 28
  6. Distinctive:

    Prefer the smallest clear expression, not a forced one-liner.

    line 30
  7. Distinctive:

    Do not invent interfaces, base classes, configuration layers, or generic helpers for one known use case.

    line 35
  8. Distinctive:

    Do not install a dependency for functionality the standard library, native platform, or existing dependency provides cleanly.

    line 37
  9. Distinctive:

    Keep the diff within the requested ownership boundary.

    line 40
  10. Distinctive:

    Ask a scope question only when the request is ambiguous or a simpler alternative materially reduces cost, risk, or ongoing maintenance.

    line 42
  11. Distinctive:

    Do not hardcode secrets, environment-specific endpoints, operational limits, or business policy merely because they have one current caller.

    line 45
  12. Distinctive:

    Add comments only for non-obvious decisions or constraints.

    line 56
  13. Distinctive:

    Add logging only when it serves an existing operational or debugging need.

    line 57
  14. Distinctive:

    Add focused verification for changed behavior, using the repository's existing test and validation conventions.

    line 58
  15. Distinctive:

    Do not add speculative retries, fallback systems, configuration, telemetry, or defensive branches for unsupported or unreachable scenarios.

    line 61
  16. Distinctive:

    Read the request, relevant code, and local conventions before editing.

    line 66
  17. Distinctive:

    Verify the behavior proportionately to its risk and blast radius.

    line 72

≈ “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

Minimalist Coding

Deliver the requested behavior with the least new complexity that remains clear, correct, and maintainable in the existing codebase. The goal is not the fewest physical lines; it is the smallest justified change.

Always respond in the user's language.

Decision Ladder

Before adding code, inspect enough of the local code and constraints to choose the first option that satisfies the request:

  1. Do not add unrequested scope - omit behavior the user did not ask for.
  2. Reuse local code - find the existing helper, pattern, or API in the codebase.
  3. Use the standard library - prefer it when it solves the problem clearly.
  4. Use the native platform - rely on framework or platform functionality already in use.
  5. Use an installed dependency - do not add a package when an existing dependency fits.
  6. Add minimal code - write only the code needed to meet the defined behavior.

Prefer the smallest clear expression, not a forced one-liner. Split code when a single expression obscures control flow, resource ownership, or failure handling.

Rules

  • Do not invent interfaces, base classes, configuration layers, or generic helpers for one known use case.
  • Do not install a dependency for functionality the standard library, native platform, or existing dependency provides cleanly.
  • Reuse the repository's established conventions before introducing a new pattern.
  • Keep the diff within the requested ownership boundary. Do not combine unrelated refactors with a feature or bug fix.
  • Ask a scope question only when the request is ambiguous or a simpler alternative materially reduces cost, risk, or ongoing maintenance. Otherwise implement the stated request.
  • Do not hardcode secrets, environment-specific endpoints, operational limits, or business policy merely because they have one current caller.

Reliability Is Proportionate To Risk

Minimalism does not mean omitting necessary safeguards.

  • Validate untrusted input at its boundary.
  • Preserve resource cleanup and existing error-propagation conventions.
  • Handle failures from I/O, networks, persistence, authentication, or concurrency when the changed path can realistically encounter them.
  • Add comments only for non-obvious decisions or constraints.
  • Add logging only when it serves an existing operational or debugging need.
  • Add focused verification for changed behavior, using the repository's existing test and validation conventions.

Do not add speculative retries, fallback systems, configuration, telemetry, or defensive branches for unsupported or unreachable scenarios.

Workflow

  1. Read the request, relevant code, and local conventions before editing.
  2. Apply the decision ladder and identify the smallest option that meets the requested behavior.
  3. State a non-obvious decision concisely when it affects scope or tradeoffs.
  4. Make the narrowest clear change; delete obsolete code rather than preserving it for hypothetical reuse.
  5. Verify the behavior proportionately to its risk and blast radius.
  6. Report what changed and the verification performed without padding the result.

Anti-Patterns

Anti-patternPreferred response
New package for a small utilityUse the standard library or an installed dependency.
Class or generic abstraction for one useKeep the logic local; extract after demonstrated reuse.
Configuration for a stable local valueKeep it local, unless it is environment-, security-, or policy-sensitive.
One-line expression that hides control flowUse a few clear lines.
Broad refactor attached to a narrow fixChange only the affected path.
Omitting likely boundary failures for brevityAdd the smallest handling consistent with local conventions.

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.