Skip to main content
subagent CX Codex

design-reviewer — Claude Code subagent: what it does, how to install

design-reviewer audits feature designs and implementation plans without conversational bias. Flags ambiguities, missing edge cases, and contradictions before implementation begins.

Elastic-2.0 ★ 155 updated 2026-10-29

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add serpro69/claude-toolbox
/plugin install kk@claude-toolbox
Open on GitHub

What design-reviewer does

The subagent inspects feature documentation, including design.md, implementation.md, and tasks.md. It remains structurally isolated from the earlier planning discussion. Because it lacks access to meeting notes or discarded alternatives, it judges specifications solely on written text.

The review workflow enforces strict sequence. It first retrieves design-review-criteria.md alongside verification-policy.md. Document evaluation never starts before these rules load. The agent then scans the repo with Read, Grep, and Glob to confirm that referenced files actually exist. Background context comes from capy_search targeting kk:arch-decisions and kk:review-findings.

The subagent lacks shell execution rights, stopping immediately if review criteria files cannot be located instead of guessing rules. It groups issues into taxonomy codes like INCOMPLETE or TECH_RISK, assigning severities from P0 to P3 and confidence scores up to 10.

What it helps with

  • spot conflicting requirements across design and task files
  • detect missing error handling or omitted migration pathways
  • verify that documented functions exist in current source code
  • evaluate architectural readiness without relying on unwritten chat context

Example prompts for the agent

  • Review design.md and tasks.md for the new billing service
  • Evaluate the storage migration implementation plan against repository patterns
  • Inspect the feature specification and list any P0 or P1 architectural concerns

How to install design-reviewer

1Plugin marketplace

/plugin marketplace add serpro69/claude-toolbox
/plugin install kk@claude-toolbox

2Manually: put the files in place

.claude/agents/design-reviewer.md

For all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention

Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean

Questions about design-reviewer

How to install design-reviewer in Claude Code?

Run /plugin marketplace add serpro69/claude-toolbox in your environment, then execute /plugin install kk@claude-toolbox to complete installation.

What does design-reviewer do?

It reads engineering planning documents, verifies their claims against the repository, and reports specification defects categorized by severity.

Why is chat history hidden from this subagent?

Omitting conversational background prevents authorship attachment. If an architectural tradeoff was discussed verbally but omitted from the document, the subagent flags it as missing.

Similar subagents

All subagents for Writing and docs →

From the same set30

The whole set →
review-architecture review-architecture is a Claude Code skill that checks a written architecture artifact against the system it describes. CX Codex +1 ★ 155repo stars review-code review-code reviews active git changes through an experienced engineering perspective. Identifies SOLID flaws, architecture issues, and security vulnerabilities. CX Codex +1 ★ 155repo stars review-design review-design is a Claude Code skill that checks design documents before implementation begins. It runs after $kk:design and reviews all documents by default. CX Codex +1 ★ 155repo stars review-spec review-spec is a Claude Code skill that compares code with feature docs and reports mismatches in both directions. Run it mid-feature or after completion. CX Codex +1 ★ 155repo stars spec-reviewer The spec-reviewer subagent checks code against design documents to find missing work, mismatches, and conflicting or unclear docs. Use it for an independent implementation review. CX Codex ★ 155repo stars eval-grader eval-grader — a grader for skill-eval assertions that returns PASS, FAIL or PARTIAL on each, with one line of evidence. Handy when a reviewer's write-up needs a verdict from something that did not produce it. CX Codex ★ 155repo stars code-reviewer code-reviewer is a Claude Code subagent that independently evaluates git diffs for SOLID principles, architectural flaws, and security risks. CX Codex ★ 155repo stars implement This skill guides planned work and standalone fixes through implementation, verification, and code review. It selects a workflow based on the request. CX Codex +1 ★ 155repo stars

Project sponsors

Companies that keep this analytics open