Automates AI-to-AI code review by bridging Claude Code and Codex CLI through Byterover shared memory, eliminating manual context switching and enabling continuous architectural validation.
git clone https://github.com/trietdeptrai/Byterover-Claude-Codex-Collaboration-.gitThis skill enables seamless collaboration between Claude Code and Codex CLI by using Byterover as a persistent shared memory layer. Claude automatically creates architectural plans, invokes Codex for expert review via Bash, retrieves feedback, and implements improvements—all without manual intervention or context switching. The workflow handles planning, automated review cycles, and optional validation in a continuous pipeline that typically completes in 5-10 minutes. Developers benefit from expert-level code review integrated directly into their AI agent workflow, catching architectural issues early and avoiding repeated mistakes across projects. This eliminates the traditional friction of manually copying context between tools.
Ensure both Claude Code and Codex CLI have Byterover MCP configured (verify with `codex mcp list`). Simply request a feature naturally, and Claude automatically creates a plan, stores it in Byterover, invokes Codex for review via Bash, retrieves feedback, and implements improvements. Optionally request validation at any point in the workflow.
Automated architectural review before implementation
Rate limiting and authentication system design
Code quality validation with expert feedback integration
Performance-critical feature implementation with dual-agent review
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/trietdeptrai/Byterover-Claude-Codex-Collaboration-Copy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
I'm working on a project for [COMPANY] in the [INDUSTRY] sector. I need to automate a workflow that involves both Claude and Codex. Can you help me design a system where Byterover serves as a shared memory layer between them? The workflow should [DESCRIBE WORKFLOW]. Provide me with a step-by-step plan, including how to set up Byterover, integrate Claude and Codex, and ensure seamless data flow between them.
# Workflow Automation Plan with Claude, Codex, and Byterover ## Overview This plan outlines a workflow automation system for [COMPANY] in the [INDUSTRY] sector, leveraging Claude, Codex, and Byterover as a shared memory layer. ## Step 1: Set Up Byterover - Create a Byterover account and set up a new project. - Define the data structure and schema for the shared memory layer. - Configure access permissions and security settings. ## Step 2: Integrate Claude - Connect Claude to Byterover using the provided API. - Set up Claude to read from and write to the shared memory layer. - Define the specific tasks Claude will handle in the workflow. ## Step 3: Integrate Codex - Connect Codex to Byterover using the provided API. - Set up Codex to read from and write to the shared memory layer. - Define the specific tasks Codex will handle in the workflow. ## Step 4: Define Workflow Logic - Outline the sequence of tasks and the conditions for triggering each step. - Define the data flow between Claude, Codex, and Byterover. - Set up error handling and fallback mechanisms. ## Step 5: Test and Deploy - Test the workflow with sample data to ensure seamless execution. - Monitor the workflow for any issues or bottlenecks. - Deploy the workflow to the production environment.
Fast food delivery from cloud kitchens
Consolidate miles from multiple cards and airlines
AI assistant built for thoughtful, nuanced conversation
Get more done every day with Microsoft Teams – powered by AI
Agentic AI Workflow platform
Connected workspace for docs, wikis, and projects
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan