← ClaudeAtlas

plan-reviewlisted

Validate an implementation plan before coding and return an evidence-backed APPROVE, REVISE, or BLOCKED verdict without rewriting it. Use for technical plan critique; use code-review for implemented diffs and fix-pr for GitHub review-comment remediation.
ValentinNikolaev/agent-skills · ★ 0 · Code & Development · score 72
Install: claude install-skill ValentinNikolaev/agent-skills
<!-- GENERATED BY generate_skill_wrappers.py; DO NOT EDIT MANUALLY. --> ## Generated Codex skill — do not edit manually This complete skill and its bundled resources are generated from `agent-plugins/skills/plan-review/SKILL.md`. Manual changes in this directory will be overwritten the next time the generator runs. Edit the canonical skill instead. # Plan Review Review an implementation plan against requirements, repository evidence, architecture, risks, and verification expectations. Do not edit files or rewrite the plan. ## Route exclusively - Use this skill only before implementation, when the primary artifact is a plan. - Use `code-review` when code has changed or the user supplies a diff. - Use `fix-pr` when the user asks to triage or apply GitHub pull-request feedback. - If no plan can be identified in the request, conversation, or supplied file, ask for it instead of inventing one. ## Choose one mode Default to `deep` unless the user requests another mode. - `quick`: verify the goal, explicit requirements, referenced artifacts, and blockers. Emit only Validation, Verdict, Blockers, Questions, and Requirements Traceability. - `deep`: also inspect applicable architecture, data flow, security, failure handling, tests, scope, and up to three useful local comparables. Emit every report section. - `final`: assume the plan is nearly ready and try to break it with production edge cases, migration and rollback failures, concurrency, observability, and recovery gaps. Emi