Index of every page on this site: llms.txt
All skills
lassejlv avatar

/no-yak-shaving

@72a388a
by Lasselassejlv/skills70 stars
2
Run it yourself

Keep implementation, refactoring, and code review proportionate to the requested behavior. Use to avoid speculative abstractions, unrelated cleanup, and tests that mirror implementation while preserving necessary contracts and validation.

Use this Skill: https://skilld.dev/gh/lassejlv/skills/no-yak-shaving

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈64 tokens for metadata: the name and description. ≈691 when used: this file. ≈59 more on demand in 1 file.

Description uses 3.2% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈255 characters, or 3.2%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

No Yak Shaving

Deliver the smallest clear solution that fully meets the requirement. Minimize unnecessary complexity, not functionality or evidence of correctness.

Choose the scope

  • Inspect the relevant call sites, contracts, and local conventions before editing.
  • Separate work required for the requested outcome from adjacent improvements. Include a prerequisite only when the outcome depends on it.
  • Reuse existing code when it fits. Prefer a direct function, explicit branch, or small data structure when no reusable mechanism is needed.
  • Preserve unrelated changes. In a review, report actionable complexity with a concrete simplification; do not turn the review into an unsolicited rewrite.

Make complexity justify itself

  • Add a layer, dependency, option, or abstraction only for a current need, such as distinct callers, a real integration boundary, or clearer ownership. A second consumer is useful evidence, not a quota; small duplication can be clearer.
  • Keep straightforward logic together unless splitting it improves navigation, ownership, or testing of meaningful behavior.
  • Preserve compatibility when callers, persisted data, or rollout requirements depend on it. Verify obsolescence before removing old behavior.
  • Validate untrusted inputs and handle failures that can occur. Omit defensive branches only when an enforced invariant makes the state unreachable.
  • Optimize performance against a measured bottleneck or an explicit target. Avoid speculative tuning and general frameworks for imagined future variants.

Validate the behavior

Use the narrowest checks that address the change's actual risk, including required repository checks. Add or update tests for a public contract, a realistic regression, consequential boundaries, or subtle branching and state transitions.

Do not add tests whose only evidence is that a private helper was called, a mock returned its configured value, or the implementation matches a copied algorithm. A simple assertion is still valuable when it protects a real contract, including a consequential default value. Preserve valuable existing tests.

For a reversible, low-impact change, existing checks or direct inspection may be enough. State what was verified and any material gap; do not claim tests ran when they did not. Do not use simplicity as a reason to skip necessary integration or user-visible verification.

Stop at completion

Before finishing, remove anything introduced by this change that can disappear without weakening the requirement, correctness, or clarity. Stop when the full requested behavior works and relevant validation is complete. Report the result and evidence briefly; mention a deferred improvement only if it matters to the user.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No alerts2mo3 checks · Risk SAFE
  • Gen Agent Trust Hub2mo

    The skill provides a set of software development guidelines focused on simplicity and avoiding over-engineering. No security issues were detected.

  • Socket2mo

    No alerts

  • Snyk2mo

    Risk: LOW · No issues

Signed by skilld at 72a388a. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 6 days ago.

Activeupdated 3 months ago

README badge

README badge for lassejlv/skills/no-yak-shaving