review-bridge

Name: review-bridge
Availability: InStock
Author: mingshi2333

Enforce a local review gate on plan artifacts before code execution.

Updated Mar 12, 2026

Search Tags:#workflow #planning #agent #code execution #review #gate

Ready to run

npx skills add https://github.com/mingshi2333/resumable-agent-workflow --skill review-bridge

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: review-bridge
Source: https://github.com/mingshi2333/resumable-agent-workflow/tree/main/claude/skills/review-bridge
Command: npx skills add https://github.com/mingshi2333/resumable-agent-workflow --skill review-bridge

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill introduces a crucial local review step before code execution, ensuring that plans and designs are formally approved or corrected, preventing costly mistakes and maintaining quality.

Core Features & Use Cases

Formal Review Gate: Integrates a reviewer subagent to inspect artifacts like plans, designs, or proposals.
Structured Decision Making: Captures review decisions (approved, changes-requested, waived) with detailed feedback.
Use Case: After a planning agent generates a proposal.md, this skill can be invoked to have a reviewer agent assess its alignment with goals and constraints before any code is written.

Quick Start

Use the review-bridge skill to start a review for the current plan.

Frequently Asked Questions about review-bridge

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I enforce a review gate for artifacts before code execution in a multi-agent workflow?▼

A local review gate enforces artifact inspection by routing documents like proposal.md to a reviewer subagent, capturing structured decisions and feedback before allowing code execution to proceed.

What is an artifact-aware review session in agent planning workflows?▼

An artifact-aware review session evaluates generated planning documents such as design.md and tasks.md, utilizing a local reviewer subagent to provide structured feedback and formal approval decisions.

How do I start a review session for a planning agent's output?▼

Invoke the review gate against the current plan to start a review session, which requires runtime integration to orchestrate the local reviewer subagent and persist the resulting review decisions.

Can I capture changes-requested decisions for design documents before generating code?▼

Yes, the review gate captures structured decisions including approved, changes-requested, and waived statuses with detailed feedback to correct or formally approve designs before code generation begins.

Do I need runtime integration to orchestrate local review gates?▼

Runtime integration is required for session orchestration and result persistence, managing the local reviewer subagent execution and storing the structured feedback generated from the artifact review.