Review Code

Run six AI code reviews in parallel and return combined findings.

Step 1: Determine the Scope

Determine what to review:

• If a specific diff command was provided (e.g., git diff --cached), use that.
• If a file list or directory was provided, review those files directly.
• If neither was provided, default to diffing against the repository's default branch (detect via gh repo view --json defaultBranchRef --jq '.defaultBranchRef.name').

Step 2: Run Six Reviews in Parallel

Launch six Agent tool calls in a single message so they run concurrently (model: "opus", do not set run_in_background). Each agent's prompt includes the scope from Step 1 and instructs it to invoke its assigned skill via the Skill tool:

• /review-test-coverage
• /review-correctness
• /review-security
• /review-quality
• /review-api-usage
• /peer-review

For the /peer-review agent, the Agent tool call prompt instructs the subagent to: (1) read the SKILL.md of every other review skill listed above, (2) extract their review criteria and "what to look for" sections, (3) compose a single comprehensive review prompt covering all dimensions with the diff command from Step 1, being verbose about what to check, and (4) invoke /peer-review via the Skill tool with the composed prompt.

Step 3: Aggregate Combined Findings

Wait for all six agents to complete. Aggregate their findings with attribution (reviewer name, file path, description).

Check your task list for remaining tasks and proceed.

Rules

• If any reviewer is unavailable or returns malformed output, proceed with findings from the remaining reviewers.
• Present findings in file order to minimize context switching.