flow-plan-review
npx machina-cli add skill gmickel/gmickel-claude-marketplace/flow-plan-review --openclawPlan Review Mode (CLI)
Conduct a John Carmack-level review of implementation plans using RepoPrompt's context builder and chat.
Role: Code Review Coordinator (NOT the reviewer) Tool: rp-cli for context building and chat delegation
Input
Arguments: #$ARGUMENTS
Format: <plan-file-or-beads-id> [additional context or focus areas]
Accepts:
- Plan file path:
plans/<slug>.md - Beads ID(s) or title(s) directly
Example: /flow:plan-review docs/plan/auth-refactor.md focus on security and error handling
FIRST: Determine Review Mode
Check: which rp-cli >/dev/null 2>&1
If NOT available: inform user rp-cli is required for this skill.
If review mode was already chosen earlier in this conversation (e.g., user answered "2a" or "2b" during /flow:plan or /flow:work setup):
→ Use that mode, don't ask again.
If invoked directly without prior context, ask:
Both modes use RepoPrompt for context building (builder, file selection, codemaps).
The difference is where the review happens:
Review mode:
a) RepoPrompt chat (default) — review via rp-cli chat
b) Export for external LLM — export context file for ChatGPT, Claude web, etc.
(Reply: "a", "b", "export", or just tell me)
Wait for response. Parse naturally.
Critical Requirement
DO NOT REVIEW THE PLAN YOURSELF – you are a coordinator, not the reviewer.
Your job is to:
- Use
rp-cli -e 'windows'to find the RepoPrompt window - Use
rp-cli -w <id> -e 'builder ...'to build context around the plan - Use
rp-cli -w <id> -e 'chat ...'to execute the review
The RepoPrompt chat conducts the actual review with full file context.
Workflow
Read workflow.md and follow each phase in order. Phases include window selection, context building, and review execution.
rp-cli Reference
Read rp-cli-reference.md for command syntax and examples.
Source
git clone https://github.com/gmickel/gmickel-claude-marketplace/blob/main/plugins/flow/skills/flow-plan-review/SKILL.mdView on GitHub Overview
flow-plan-review orchestrates a rigorous assessment of implementation plans, architecture docs, or design specs. It uses RepoPrompt context building and chat via rp-cli, acting as the Code Review Coordinator (not the reviewer) to ensure thorough, traceable feedback.
How This Skill Works
As the coordinator, you don’t review directly. The skill checks rp-cli availability, selects a review mode, builds context around the plan with RepoPrompt builder, then executes a chat-based review through rp-cli. The actual critique is performed by the RepoPrompt chat with full file context.
When to Use It
- Review a plan file (plans/*.md) for correctness, risks, and completeness.
- Assess architecture diagrams and design specs against requirements.
- Validate security, error handling, and scalability considerations.
- Export context for external LLMs when you want a standalone review artifact.
- Coordinate reviews across teams without performing the review yourself.
Quick Start
- Step 1: Ensure rp-cli is installed and decide the review mode (a, b, or export).
- Step 2: Provide the plan path (plans/<slug>.md) or beads and run the builder to assemble context.
- Step 3: Run the chat-based review with rp-cli and review the output.
Best Practices
- Verify rp-cli is available before starting the workflow.
- Provide precise focus areas (e.g., security, API surface, data flow) to guide the review.
- Treat the Code Review Coordinator role as responsible for flow control, not critique.
- Use the builder step to gather relevant files, codemaps, and context around the plan.
- Document actionable findings with risk levels and owners for follow-up.
Example Use Cases
- Review a plan for a microservices refactor focusing on module boundaries and testing strategy.
- Assess an architecture diagram for a cloud-native deployment with networking and tenancy concerns.
- Critique an API design spec to ensure backward compatibility and clear error semantics.
- Security review of an authentication flow and data protection in a new plan.
- Coordinate a cross-team design review for a new service with multiple dependencies.
