code-review

Runs senior-level code review on feature pull requests within a spec-driven development pipeline.

Updated Aug 25, 2026
One-click install
npx skills add https://github.com/asengardeon/btt-sdd-pipeline --skill code-review-asengardeon
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/asengardeon/btt-sdd-pipeline/tree/main/plugins/btt-sdd/skills/code-review
Command: npx skills add https://github.com/asengardeon/btt-sdd-pipeline --skill code-review-asengardeon

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After a feature is implemented and a pull request is open, teams need a rigorous senior-engineer review covering ports & adapters architecture, SOLID principles, clean code, and test quality before QA begins. This Skill orchestrates that review stage so no verification step is silently skipped. ## Core Features & Use Cases - Agent-Orchestrated Review: Invokes the code-reviewer agent with worktree isolation to produce a structured review report at specs/<slug>/code-review.md from a template. - Branch Synchronization Checks: Verifies the feature branch is rebased on the latest main before reviewing, with conflict-resolution guidance for spec artifact files. - Pipeline Governance: Logs invocation timing to timing-log.md, routes rejected work back to implementation, and advances approved work to the QA stage. - Use Case: A developer finishes implementing a feature on branch feature/0042-checkout, opens a PR, and runs this stage to get a senior-level verdict (approved, approved with reservations, or rejected) with documented findings before QA testing starts. ## Quick Start Ask the assistant to run the code review stage for the feature whose pull request was just opened, passing the feature slug and PR reference.

Frequently Asked Questions about code-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a code review on a feature pull request?▼

Invoke this stage after implementation is complete and a PR is open on a feature/<NNNN-slug> branch. It launches the code-reviewer agent with the TRD path and PR reference, producing specs/<slug>/code-review.md with a verdict and findings.

What does the code review check in a ports and adapters architecture?▼

The review covers ports and adapters structure, SOLID principles, clean code practices, and test quality. The reviewer reads the actual PR diff before issuing any verdict of approved, approved with reservations, or rejected.

What happens when a code review is rejected?▼

The feature returns to the implementation stage with the listed findings. The recommended path is resuming the same implementing agent via SendMessage for small targeted fixes rather than starting a new agent.

Why must the feature branch be rebased before code review?▼

Rebasing on the latest main prevents later stages from committing onto a conflicting base. If the rebase brought substantial commits from another merged feature, the reviewer must re-verify propagation sweeps that were run before the rebase.

Can I run the code review without the agent tool?▼

Yes. If the Agent tool is unavailable, follow the same process described in the code-reviewer agent definition directly, reading the PR diff yourself before giving any verdict.