What problem does it solve? Code reviews often conflate two separate questions — is the code well-written, and does it do what was asked — letting one axis mask failures in the other. This Skill separates them into independent parallel reviews so standards violations and spec gaps are each visible on their own. ## Core Features & Use Cases - Two-axis review: Runs a Standards review (repo conventions plus the twelve-smell Fowler baseline) and a Spec review (requirements coverage, scope creep, wrong implementations) as parallel sub-agents that don't pollute each other's context. - Fixed-point diffing: Compares HEAD against any user-supplied commit, branch, or tag using three-dot merge-base diffs, so you can review a branch, a PR, or work-in-progress since any point. - Automatic source discovery: Locates the originating spec from issue references in commit messages, user-provided paths, or docs/specs files, and collects standards sources like CLAUDE.md, CONTRIBUTING.md, ADRs, and lint configs. - Use Case: Before merging a feature branch, ask for a review against main and receive separate Standards and Spec reports with cited rules, quoted spec lines, and a one-line summary of the worst finding. ## Quick Start Ask the agent to review the current branch against main, for example: review my changes since main and check them against our coding standards and the original issue.
