← Browse

@goniz/mlx-review

A

Review changes in the MLX submodule for concrete bugs, behavioral regressions, structural mismatches, and obvious performance problems. Use when asked to review uncommitted `mlx/` changes, a supplied diff, commit, file, or change set and provide actionable evidence-based feedback without modifying code.

skillclaude

Install

agr install @goniz/mlx-review --target claude

Writes 1 file into .claude/skills/, pinned to git-3073f72e.

  • .claude/skills/mlx-review/SKILL.md

Document


name: mlx-review description: Review changes in the MLX submodule for concrete bugs, behavioral regressions, structural mismatches, and obvious performance problems. Use when asked to review uncommitted mlx/ changes, a supplied diff, commit, file, or change set and provide actionable evidence-based feedback without modifying code.

Review MLX Changes

Review code changes, not pre-existing code. Do not modify files.

  1. Resolve the requested review target. With no target, inspect git -C mlx diff, git -C mlx diff --cached, and git -C mlx status --short.
  2. Read the full contents of every modified and untracked file, not only the diff. Read AGENTS.md, .editorconfig, and other applicable conventions.
  3. Investigate uncertain behavior through surrounding code and established patterns before reporting an issue.
  4. Prioritize concrete bugs: logic and boundary errors, incorrect guards, error handling, race conditions, security risks, and unintended behavior changes.
  5. Flag structural mismatches only when they conflict with established project patterns. Flag performance only when clearly problematic, such as unbounded quadratic work or blocking I/O on a hot path.
  6. Do not invent hypothetical concerns. State the realistic input, environment, or control-flow condition required for every finding.

Report findings first, ordered by severity, with path:line, a concise explanation, the triggering scenario, and why it is a defect. State explicitly when no actionable findings remain. Keep the tone factual and avoid non-actionable praise or style preferences.

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-3073f72e6b482026-07-31