Plan Review
Scannednpx machina-cli add skill smith-horn/product-builder-starter/plan-review-skill --openclawPlan Review
Multi-perspective plan review with VP Product, VP Engineering, and VP Design to identify blockers, anti-patterns, conflicts, and regressions.
Execution
When triggered, immediately:
- Read
agent-prompt.mdfrom this skill's directory:- User install:
~/.claude/skills/plan-review-skill/agent-prompt.md - Project install:
.claude/skills/plan-review-skill/agent-prompt.md
- User install:
- Spawn a single Task with
subagent_type: "general-purpose"passing the agent-prompt content as the prompt - Include in the prompt: the user's request, the plan file path or content, and current working directory
- Wait for the agent to complete
- Present the agent's summary to the user
Do NOT execute the review workflow in this session. The subagent handles VP agent spawning, issue consolidation, and user interaction (AskUserQuestion). Plan edits are returned as recommendations for the coordinator to apply — the subagent should not edit files directly, as background execution auto-denies Write/Edit tools.
Execution Context Requirements
This skill spawns a general-purpose subagent that coordinates VP review agents and consolidates findings.
Foreground execution required: Yes for interactive AskUserQuestion prompts. Background execution auto-denies clarifying questions.
Dispatcher tools (frontmatter): Read, Task Subagent tools: Read, Grep, Glob, AskUserQuestion
Write/Edit delegation: The subagent returns approved plan modifications as text. The coordinator (main conversation) applies edits via its own Edit tool. This makes the skill resilient to all permission modes.
Reference: See Claude Code documentation on subagent tool permissions.
Changelog
v2.0.0
- Refactor: Thin dispatcher pattern — full logic extracted to agent-prompt.md
- Skill runs in isolated subagent context, significantly reducing post-compaction restoration overhead
See agent-prompt.md for prior changelog entries.
Source
git clone https://github.com/smith-horn/product-builder-starter/blob/main/skills/plan-review-skill/SKILL.mdView on GitHub Overview
Plan Review brings VP Product, VP Engineering, and VP Design together to assess plans such as implementation plans, design docs, architecture proposals, or wave plans. It focuses on blockers, anti-patterns, conflicts, and regressions to improve risk mitigation and stakeholder alignment.
How This Skill Works
When triggered, the skill reads agent-prompt.md and spawns a general-purpose subagent with that prompt. The subagent receives the user request, the plan file path or content, and the current working directory, then consolidates findings and returns a summary of recommendations; the coordinator presents this to the user. Edits are returned as recommendations, not direct changes; the coordinator applies edits via its own Edit tool.
When to Use It
- Review an implementation plan for blockers or risks.
- Evaluate a design document for architectural anti-patterns.
- Assess an architecture proposal for conflicts across teams.
- Check a wave plan for potential regressions when syncing releases.
- Prepare for executive or stakeholder review before funding or launch.
Quick Start
- Step 1: Trigger Plan Review with your plan file path or content and any context.
- Step 2: The system reads agent-prompt.md and spawns a general-purpose subagent.
- Step 3: Review the subagent’s summary and apply approved edits via the coordinator’s Edit tool.
Best Practices
- Provide the plan’s file path or content and the context in your prompt.
- Spell out explicit success criteria and known blockers to guide the review.
- Redact sensitive data before submitting the plan for review.
- Ask the subagent to consolidate issues and present actionable recommendations.
- Review the recommended edits with the team and apply changes using the coordinator’s Edit tool.
Example Use Cases
- Review a feature implementation plan to surface blockers and alignment gaps.
- Audit a microservices redesign proposal for cross-team conflicts.
- Assess a wave plan to identify regressions across release trains.
- Check a design document for UX- and tech-debt alignment with architecture goals.
- Coordinate VP-level approval by surfacing risks and recommended next steps.
Frequently Asked Questions
Related Skills
compose
chaterm/terminal-skills
Docker Compose 编排
unified-review
athola/claude-night-market
'Use this skill when orchestrating multiple review types. Use when general
Hive Workers
smith-horn/product-builder-starter
Execute tasks using hive mind orchestration with parallel agents, automatic code review, and documentation updates. Use for feature execution, epic completion, or complex multi-task work.
flows
victorgrein/spec-crew
This skill should be used when designing CrewAI Flows with typed state management, deterministic routing, and event-driven orchestration. Use it for flow architecture, state transitions, control primitives, and production deployment.
