Skip to content
SkillModel-led662 tokens in use

requesting-code-review

Sends a fresh code reviewer subagent the diff and requirements, then sorts findings into Critical, Important and Minor.

322,762 stars on the repository MIT Licenseobra/superpowers · skills/requesting-code-review
/plugin install superpowers@claude-plugins-official
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing requesting-code-review costs

22tokens idle
662tokens in use
2,114tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. 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

shelldetectedgitdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. 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
  • Structure14/25
  • Examples12/15
  • Output10/15
  • Size13/15
  • Resources15/15

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
  • Runs shell commands−5

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 28

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity56
  • Maintenance68
  • Community64
  • Maturity36
  • Momentum40

Requesting Code Review has your agent dispatch a reviewer subagent after each task, after a major feature and before merging. It collects the base and head commit SHAs, fills in a reviewer template with a description and the requirements, and gets back strengths, issues by severity and an assessment. Critical issues are fixed at once, Important issues before moving on, and Minor ones are noted for later.

What You Get

  • A trigger list: mandatory after each task, after major features and before merging to main
  • A reviewer template with placeholders for description, requirements and commit range
  • A severity scheme for acting on findings
  • A worked example of a review after a task

Common Use Cases

  • Reviewing each task in a subagent-driven run
  • Checking a finished feature before merging to main
  • Getting a fresh perspective when stuck or before a refactor

What Makes It Unique

The review runs in a separate subagent’s context, so the diff and its evaluation stay out of the coordinator’s window and only the findings come back. The skill names the rationalizations that lead agents to skim the diff themselves and answers each one.

Under The Hood

Instructions plus a code-reviewer.md template. The agent runs git rev-parse and git merge-base to get commit SHAs and dispatches a general-purpose subagent with the filled template. It runs nothing else and calls no outside services.

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