flow-impl-review
Scannednpx machina-cli add skill gmickel/gmickel-claude-marketplace/flow-impl-review --openclawImplementation Review Mode (CLI)
Conduct a John Carmack-level review of implementation changes on the current branch 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: [additional context, focus areas, or special instructions]
Example: /flow:impl-review focus on the auth changes, ignore styling
Reviews all changes on the current branch vs main/master.
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.
Context Sources
The workflow gathers context from:
- Git diff and commit messages
- Plan files referenced in commits
- PRD/architecture docs
- Beads: If you know which issue(s) this work relates to, include via
bd show
Critical Requirement
DO NOT REVIEW CODE 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 - 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 change identification, 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-impl-review/SKILL.mdView on GitHub Overview
flow-impl-review coordinates a deep, Carmack-level implementation review for the current branch using RepoPrompt and rp-cli. As the Code Review Coordinator, you orchestrate the process, not perform the actual code critique, and you guide context gathering, mode selection, and review execution.
How This Skill Works
You first determine if rp-cli is available; if not, you inform the user. You then decide on a review mode (default chat or export for external LLM) and use rp-cli to locate the RepoPrompt window, build the review context, and finally run the actual review via RepoPrompt chat. The workflow relies on collecting diffs, plan files, and docs to produce a comprehensive review without touching the code yourself.
When to Use It
- Reviewing updates on the current feature branch before merging to main/master
- Auditing PRs with high-risk implementation changes (security, performance, architecture)
- Coordinating a review workflow where another reviewer performs the actual critique
- Preparing and exporting context for external LLMs (e.g., ChatGPT, Claude) as part of the review
- Responding to mentions of code review, implementation review, or PR review within a project
Quick Start
- Step 1: Verify rp-cli is installed and locate the RepoPrompt window
- Step 2: Decide the review mode (a for RepoPrompt chat, b or export for external LLM) and respond accordingly
- Step 3: Build context with rp-cli using rp-cli -e 'windows', rp-cli -w <id> -e 'builder ...', then rp-cli -w <id> -e 'chat ...'
Best Practices
- Define the exact focus areas (diffs, architecture, risk) before starting
- Decide the review mode upfront (a: RepoPrompt chat, b: export for external LLM)
- Use rp-cli to locate the RepoPrompt window and build complete context
- Never review code yourself; act as the coordinator and delegate the actual critique
- Document outcomes and preserve context artifacts for audit or export
Example Use Cases
- Coordinating a Carmack-level review of feature/login branch changes before PR merge
- Reviewing a performance-critical refactor in module Y and summarizing risks
- Auditing a security patch across multiple files and generating an export for LLM tooling
- Preparing context for a large PR so an external LLM can assist in the review
- Handling a PR review request with explicit mentions of code and implementation review
