Is specific software required to use this skill?

Yes, you must have either rp-cli (RepoPrompt) or codex (Codex CLI) installed and configured on your system for the skill to perform the actual review analysis.

Why use an external backend for code reviews?

Using external backends like RepoPrompt or Codex prevents the AI from 'marking its own homework,' providing a more objective, high-tier evaluation of code changes.

How does the 'fix-and-re-review' loop operate?

If the review returns a 'NEEDS_WORK' verdict, the skill automatically parses the feedback, applies fixes, commits them, and re-triggers the review until a 'SHIP' verdict is achieved.

Can this skill be used in fully autonomous workflows?

Absolutely. It is specifically designed to work with Ralph autonomous mode, acting as a quality gate that prevents broken or sub-par code from being merged during overnight sessions.

Flow-Next Implementation Review

Name: Flow-Next Implementation Review
Author: gmickel

bygmickel

•

356

Collaboration Tools

Coordinates expert-level code reviews using RepoPrompt or Codex to ensure production-grade implementation quality and architectural integrity.

About

The flow-next-impl-review skill transforms Claude Code into a sophisticated review coordinator that interfaces with high-tier analysis backends like Codex CLI and RepoPrompt. Instead of relying on a single-pass self-review, it enforces a 'John Carmack-level' scrutiny process that evaluates branch diffs against the main codebase. It manages the entire lifecycle of a code review—from backend selection and prompt building to an internal 'fix-and-re-review' loop—ensuring that code changes meet rigorous standards before receiving a final 'SHIP' verdict.

Key Features

Internal fix-and-re-review loops for rapid issue resolution
Receipt-based session continuity for persistent review contexts
356 GitHub stars
Seamless integration with git branch diffs and Ralph autonomous mode
Multi-backend coordination for RepoPrompt and Codex CLI
Automated 'John Carmack-level' implementation analysis

Use Cases

Enforcing high-quality review gates for complex Pull Requests
Standardizing elite engineering reviews across diverse development teams
Autonomous overnight coding sessions requiring external validation

About

Key Features

Internal fix-and-re-review loops for rapid issue resolution
Receipt-based session continuity for persistent review contexts
356 GitHub stars
Seamless integration with git branch diffs and Ralph autonomous mode
Multi-backend coordination for RepoPrompt and Codex CLI
Automated 'John Carmack-level' implementation analysis

Use Cases

Enforcing high-quality review gates for complex Pull Requests
Standardizing elite engineering reviews across diverse development teams
Autonomous overnight coding sessions requiring external validation