Saucebox Registry
← Browse

@diegopacheco/agent-skill-linter

A

Lint a codebase on code quality, best practices, design principles, build and test status, per-test speed (slow at or above 5s), cyclomatic complexity, and expressive naming, then render the results as a modern web report. Use when the user runs /lint or /lint-site, or asks to lint, score, or audit the quality of a project.

skillclaude

Install

agr install @diegopacheco/agent-skill-linter --target claude

Writes 1 file into .claude/skills/, pinned to git-ced8eaf6.

  • .claude/skills/agent-skill-linter/SKILL.md

Document

name: agent-skill-linter description: Lint a codebase on code quality, best practices, design principles, build and test status, per-test speed (slow at or above 5s), cyclomatic complexity, and expressive naming, then render the results as a modern web report. Use when the user runs /lint or /lint-site, or asks to lint, score, or audit the quality of a project.

agent-skill-linter

A hybrid linter. Deterministic signals come from host tooling; semantic signals come from your judgment. The two halves merge into one report.json.

Layout (after install)

  • ~/.claude/skills/agent-skill-linter/engine/lint.mjs — deterministic analyzer
  • ~/.claude/skills/agent-skill-linter/engine/assemble.mjs — merges deterministic + semantic into the final report
  • ~/.claude/skills/agent-skill-linter/site/ — the web report stack (Podman)
  • commands /lint and /lint-site

Output (written into the target repo)

  • .lint/deterministic.json — produced by the engine
  • .lint/semantic.json — produced by you (the model) during /lint
  • .lint/report.json — merged final report
  • .lint/history/<timestamp>.json — one trimmed entry per run, for trends

Scored categories and weights

build 20, tests 20, complexity 15, principles 15, bestPractices 12, codeQuality 10, naming 8.

  • Deterministic: build, tests (including the slow flag at 5s), complexity (cyclomatic per function), plus function-length and file-length checks under codeQuality.
  • Semantic (your judgment): naming, principles (SOLID, DRY, KISS, separation of concerns), bestPractices, and overall codeQuality.

How to run

/lint [path] then /lint-site [path]. See the two command files for the exact steps.

Trustgrade A

  • passBody integrity

    Whether the stored document is plausibly the kind of file the artifact declares, rather than something fetched by mistake.

  • passType matchnot applicable to this artifact type

    Whether the artifact is really the kind of thing its metadata claims it is.

  • passFreshness

    How long since the source repository was last pushed to.

  • passPrompt injection

    Scans the artifact's own text for instructions aimed at your agent rather than at you.

  • passLicense

    Whether the source repository declares an SPDX license permissive enough to redistribute.

How the grade is calculated

Each check contributes 0 points when it passes, 1 when it warns, and 2 when it fails. The total maps to a letter:

  • Aevery check passed
  • Bone warning
  • Ctwo warnings
  • Dprompt injection or body integrity failed, or three warnings
  • Fone of those failed, and something else is wrong

These are automated hygiene checks, not a security audit, and not a dependency or vulnerability scan. A grade of A means nothing was flagged — not that the artifact is safe.

Versions

  • git-ced8eaf6a2b62026-07-31