ln-12-delivery-reviewer

Reviews completed code changes against acceptance criteria and produces evidence-based verdicts.

556|83|Updated Oct 24, 2025
One-click install
npx skills add https://github.com/levnikolaevich/claude-code-skills --skill ln-12-delivery-reviewer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ln-12-delivery-reviewer
Source: https://github.com/levnikolaevich/claude-code-skills/tree/main/plugins/review-suite/skills/ln-12-delivery-reviewer
Command: npx skills add https://github.com/levnikolaevich/claude-code-skills --skill ln-12-delivery-reviewer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Verifying that a finished code change actually delivers its intended business outcome is hard: reviewers often miss regressions, accept unproven claims, or drift into unrelated code audits. This Skill enforces a scoped, evidence-driven delivery review that proves or disproves release readiness.

Core Features & Use Cases

  • Scoped Change Review: Traces every task requirement and acceptance criterion to concrete implementation and behavioral evidence, marking each COMPLETE, DEVIATED, OMITTED, or UNPROVEN.
  • Independent Review Panel: Coordinates Six Thinking Hats lenses plus risk-triggered specialists (security, data, API, tests, UI, operations) as read-only subagents within a strict two-round budget.
  • Structured Verdict Report: Produces a PASS, CONCERNS, FAIL, or BLOCKED verdict with prioritized findings (P0-P3), test and documentation actions, and residual risks.
  • Use Case: After a developer finishes a feature branch, run this Skill before merge to verify acceptance criteria, detect change-caused defects, check contract compatibility, and confirm tests and docs are correct.

Quick Start

Review the completed changes on this branch against the task requirements and acceptance criteria, then give me a verdict with evidence.

Frequently Asked Questions about ln-12-delivery-reviewer

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

FAQPage Schema
How do I review a code change before merging?

Provide the task requirements, acceptance criteria, and the base and head of the change. The review traces each requirement to implementation and behavioral evidence, runs available build and test gates, and returns a verdict with prioritized findings.

What is the difference between delivery review and a code audit?

Delivery review examines only the requested change and the causal paths needed to prove its business outcome. It does not audit unrelated code, report general repository health, or repair findings; those belong to codebase audit work.

Can this review use subagents for independent verification?

Yes. It supports an independent panel using Six Thinking Hats lenses plus risk-triggered specialists such as security, data, API, and UI reviewers. The budget is at most two rounds: one initial scope-scaled review and one selective follow-up.

When should I not use delivery review?

Do not use it for full codebase audits, implementing fixes, or repairing findings. It is read-only by design and only judges scoped acceptance and release safety for a completed change.

What verdicts does the delivery review produce?

It produces PASS, CONCERNS, FAIL, or BLOCKED. FAIL applies to unresolved P0/P1 findings or unmet acceptance, BLOCKED to unproven required evidence, CONCERNS to non-blocking risk, and PASS only when all required evidence is complete.