Skip to content
Agent Memory
Command

/review

Before dispatching the reviewer, determine the correct working directory:

BOOST
From plugin
oh-my-kiro
10715 skills2 agents15 commands2 hooks
Install
$ npx -y skills add kaimingwan/oh-my-kiro --agent claude-code

How it fires

How this command gets triggered: by you, by Claude, or both.

  • Fires itselfClaude auto-loads it when your prompt matches the work.
  • You can call itInvoke it directly when you want it.
  • Slash command/review

Context preview

What this command does when you run it.

Before dispatching the reviewer, determine the correct working directory:

Command definition

review.md

Before dispatching the reviewer, determine the correct working directory:

Step 1: Resolve review target

1. If the user specifies a path after @review (e.g., `@review worktrees/omk-foo`), use that path. 2. Otherwise, check if `.active-submodule` exists in the project root:

if [ -f .active-submodule ]; then
  jq -r '.worktree // empty' .active-submodule
fi

3. If a worktree path is found, use it as the review working directory. 4. If neither is available, review the project root (default behavior).

Set the resolved path as `REVIEW_DIR`.

Step 2: Gather diff context

Run in the resolved directory to build the review query:

cd "$REVIEW_DIR"
git diff --stat
git diff

If the diff is empty (no unstaged changes), also check staged changes:

cd "$REVIEW_DIR"
git diff --cached --stat
git diff --cached

Step 3: Dispatch reviewer

Dispatch a reviewer subagent (`agent_name: "reviewer"`) with this query:

"Review the code changes in `<REVIEW_DIR>`. Run `git -C <REVIEW_DIR> diff --stat` then `git -C <REVIEW_DIR> diff`. If no unstaged changes, check `git -C <REVIEW_DIR> diff --cached`. Categorize findings: P0 Critical / P1 High / P2 Medium / P3 Low. Check: correctness, security, SOLID violations, test coverage, edge cases. Be specific — cite file:line and show code examples."

Report the findings to me.

Ships withoh-my-kiro

Your AI agent forgets everything. Mine doesn't. oh-my-kiro is a framework that gives your AI coding agent persistent memory, deterministic workflows, and self-evolving intelligence. Like oh-my-zsh for Zsh — but for AI agents. Day 1: generic agent.

Get the whole plugin

Other commands on oh-my-kiro.