multi-reviewer-patterns Skill (wshobson) for Claude: What It Does and How to Install It

Makes Claude split a code review into dimensions, merge duplicate findings and rate severity the same way every time.

Installs: 10K; on skills.sh
Stars: 42.1K; on GitHub
Files: 1; besides SKILL.md
Updated: Oct 2026; last repo commit

This skill is a playbook for code reviews that look at several things at once: security, performance, architecture, testing and accessibility. It tells Claude which dimensions to pick for a kind of change, how to merge findings that point at the same line, and how to grade them Critical, High, Medium or Low. It ends with a report template.

It triggers when you organize a multi-reviewer review, calibrate severity or consolidate results. It bundles no scripts and has one checklist file. It is part of wshobson's agent-teams plugin, whose commands use Claude Code's experimental Agent Teams feature, but the skill text itself does not call any of them.

Good for

Developers running a multi-angle review of a pull request
Tech leads who want consistent severity ratings across reviewers
Teams using several reviewer agents in parallel

Install multi-reviewer-patterns

Claude Code

Copy this prompt.

Install the multi-reviewer-patterns skill in Claude Code. Run this command:

/plugin marketplace add wshobson/agents
/plugin install agent-teams@claude-code-workflows

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Pick one route. The plugin routes install the whole agent-teams plugin: 4 agents, 7 commands and 5 skills (counted from the plugin folder in the repo tree, which has no hooks or MCP config files). The plugin's README says its commands need Claude Code's experimental Agent Teams feature, turned on with the CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS environment variable. The skills-only routes install just this skill, with no agents or commands.

From the project README. Source.

What's in the skill

When it loads

Coordinate parallel code reviews across multiple quality dimensions with finding deduplication, severity calibration, and consolidated reporting. Use this skill when organizing multi-reviewer code reviews, calibrating finding severity, or consolidating review results.

Sections in SKILL.md

When to Use This Skill
Review Dimension Allocation
Available Dimensions
Recommended Combinations
Finding Deduplication
Merge Rules
Deduplication Process
Severity Calibration
Severity Criteria
Calibration Rules
Consolidated Report Template
Code Review Report
Critical Findings ({count})
High Findings ({count})
Medium Findings ({count})
Low Findings ({count})

Key facts

Author: wshobson (repo owner)
License: MIT
Installs on skills.sh: 9,974 (Nov 2, 2026)
GitHub stars: 42,081 for the whole wshobson/agents repo (Nov 2, 2026)
Last repo commit: Oct 30, 2026
Audits on skills.sh: Gen Agent Trust Hub: pass, Socket: pass, Snyk: pass
What it includes: 2 files including SKILL.md: references/review-dimensions.md (no scripts)
Part of: The agent-teams plugin in wshobson/agents. Counted from the plugin folder: 4 agents, 7 commands and 5 skills

What it includes

Picks from 5 dimensions and matches them to the change
A table of security, performance, architecture, testing and accessibility, plus recommended combinations such as API endpoint changes or database migrations.
Try: “Which review dimensions should I assign for this auth change?”
Read · Review dimension allocation
Merges findings that hit the same file and line
Five merge rules, including: same issue at the same place becomes one finding, and the higher severity wins.
Read · Finding deduplication
Grades findings Critical, High, Medium or Low
A criteria table plus rules such as: externally exploitable security issues are always Critical or High.
Read · Severity calibration
A report layout with findings by severity and a summary table
Fill-in markdown for target, reviewers, findings per severity, a per-dimension table and a recommendation.
Write · Consolidated report template
Per-dimension checklists in references/review-dimensions.md
Checklists for security, performance, architecture, testing and accessibility.
Read · Dimension checklists

From the official docs

Permissions

Low risk

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

The plugin route adds agents with Bash access
The agent-teams plugin has 4 agents. Their tool lists include Bash, and the team-implementer agent also has Write and Edit. Use the skills-only route if you only want the review rules. github.com
Plugin commands need an experimental feature
The plugin README says to set CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1. The skill itself does not need it. github.com

Security advisories

No published security advisories as of Nov 2, 2026.

Technical details

Turning on read-only	No read-only mode. Use version control.
Sign-in and keys	SKILL.md and the reference file do not mention any account or key.
Scopes	Reads the code you point it at and can write a review report when asked

Questions

What does the multi-reviewer-patterns skill do?

It gives Claude rules for running a code review across several dimensions. It covers which dimensions to pick, how to merge duplicate findings, how to rate severity and how to format the final report.

How do I install multi-reviewer-patterns?

Use the install box at the top of this page. It offers the whole agent-teams plugin for Claude Code and Codex, and skills-only routes that install just this skill.

Do I need the agent-teams plugin?

No. The skill text does not call any plugin command or agent, so the skills-only routes are enough for the review rules. The plugin adds 4 agents, 7 commands (such as /team-review) and 4 more skills, and its README says the commands need Claude Code's experimental Agent Teams feature.

Is this skill from Anthropic?

No. It is in wshobson/agents, a community collection, inside the agent-teams plugin.

Is it safe?

It bundles no scripts, only SKILL.md and one checklist file. The guidance can lead Claude to read your code and write a review report, so check any file it creates. The plugin route adds agents whose tool lists include Bash and, for the implementer agent, Write and Edit.

How is it different from /team-review?

/team-review is a plugin command that runs a parallel review. This skill holds the rules for dimensions, deduplication and severity that such a review follows. The skill works without the command.

Information

updated Nov 2, 2026

Repository: wshobson/agents
Folder: plugins/agent-teams/skills/multi-reviewer-patterns

wshobson/agents skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 5

multi-reviewer-patterns SKILL.md · github.com
review-dimensions.md · github.com
wshobson/agents README · github.com
agent-teams plugin.json · github.com
agent-teams plugin README · github.com