← ClaudeAtlas

add-plan-reviewlisted

Pre-delivery executability review for freshly generated about.md and plan.md — verdict (ok / fix-then-ok / blocked) plus required fixes, not questions. Load before dispatching plan-reviewer-agent, or when applying its rubric inline as a fallback.
brabos-ai/code-addiction · ★ 9 · Code & Development · score 76
Install: claude install-skill brabos-ai/code-addiction
# Plan Review <!-- uses: - mention: add-code-review - mention: add-doc-schemas - mention: add-feature-readback - mention: @ux-agent - agent: plan-reviewer-agent - agent: reviewer-agent - command: /add.new - command: /add.plan --> ## Overview Rubric and verdict contract for the fresh-reader, pre-delivery review that `add.new` and `add.plan` run after their validation gate passes and before the doc is presented as delivered. The reviewer tries to break the document — find what would fail in execution — and returns a verdict with concrete, evidence-backed fixes. This is a **fix-oriented** review: it proposes the fix, not just the question. A review that only asks hands the work back; this one hands back a verdict and the concrete edit that clears it. ## When to Use - `plan-reviewer-agent` loads this skill as its rubric source - `/add.new` (STEP 8, full path only) and `/add.plan` (STEP 12) dispatch the agent, or apply this skill inline as a fallback when the provider does not support subagent dispatch ## When NOT to Use - Code review — use `add-code-review` / `@reviewer-agent`. No code exists at this phase. - Schema-compliance enumeration (frontmatter shape, required sections, depth floors) — that is the validation gate inside `add-doc-schemas`. This review runs strictly after that gate passes. - A replacement for `@ux-agent` critique mode — `design.md` already has its own adversarial reviewer inside `/add.plan` STEP 7.1. - Questioning without a proposed fix. A finding tha