Skip to content
SkillModel-led1,426 tokens in use

Code Review

Reviews the changes since a fixed point along two axes, repo standards and spec match, with parallel sub-agents and a Fowler smell baseline.

328,754 stars on the repository MIT Licensemattpocock/skills · skills/engineering/code-review
claude plugin install mattpocock-skills@claude-plugins-official
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing Code Review costs

94tokens idle
1,426tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts and code blocks. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Good
  • Frontmatter12/15
  • Structure23/25
  • Examples3/15
  • Output10/15
  • Size15/15
  • Resources0/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity96
  • Maintenance100
  • Community56
  • Maturity32
  • Momentum40

Code Review compares HEAD against a commit, branch, tag or merge-base that you supply and reports on two axes: Standards (does the code follow this repo’s documented coding standards) and Spec (does it match what the originating issue or spec asked for). Each axis runs in its own sub-agent with a fresh context window.

The two reports appear side by side under separate headings and are never merged or reranked.

What You Get

  • A Standards report listing violations of documented standards, each citing the standard’s file and rule
  • A Spec report listing missing or partial requirements, scope creep and requirements that look wrongly implemented, each quoting the spec line
  • A smell baseline of twelve Fowler code smells, such as Duplicated Code, Feature Envy and Shotgun Surgery, flagged as judgement calls
  • A closing one-line summary of findings per axis and the worst issue within each axis

Common Use Cases

  • Reviewing a feature branch against its originating issue before opening a pull request
  • Checking work in progress with a review since a commit, tag or main
  • Closing out an implementation run, since implement and implement-spec both call this skill

What Makes It Unique

Standards and Spec are separate axes with separate sub-agents, so code that follows every convention but builds the wrong thing is caught, and so is the reverse. The skill reports the two sets of findings without merging or reranking them. A documented repo standard always overrides the built-in smell baseline.

Under The Hood

The skill is Markdown instructions that run git commands: git rev-parse to check the ref, git diff with the three-dot form, and git log for the commit list. It fetches issues through the tracker workflow that the setup skill configures, reads files in the repo, and starts two sub-agents in the foreground. It writes no files.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers