On their own
- Write the change in your project folder
- Run your tests
- Review the change on a second engine
- Hand the fixes back to the builder
Founders and leadership
Have Codex review what Claude Code built.
Put one bot on Claude Code to build and another on Codex to review. Your Chief of Staff hands the finished change from one to the other, so a second model reads every change before you merge it.
Free and open source. macOS, Windows and Ubuntu.
Engine review
Have the builder add CSV export to the reports page. When it’s done, hand it to the reviewer on Codex and tell me what it would change. Merge nothing.
Your Chief of Staff put a team on it
- Ops
- Review
CSV export: built and reviewed
- Builder on Claude Code: 4 files changed, tests pass
- Reviewer on Codex: 2 issues, 1 suggestion
- Issue: dates export in UTC, not the user’s time zone
- Issue: no limit on export size
Needs your answer
Hand the two issues back to the builder?
An example of the job. Your team works on your own work.
The team it puts on it.
Your Chief of Staff picks the specialists the job needs. Each one has its own role, instructions and engine.
Chief of Staff
Hands the build to one bot and the review to the other, then reports back
Ops
The builder: a bot on Claude Code that writes the change and runs your tests
Review
The reviewer: a bot on Codex that reads the change fresh and lists what it would fix
How it runs.
- Step 1An engine per botEach bot runs on the engine you pick for it. Put the builder on Claude Code and the reviewer on Codex, the official Codex CLI you install and sign in to yourself.
- Step 2Give the team your repoSet the team’s working folder to your project, so both bots read the same code.
- Step 3Build, then hand offThe builder makes the change. The Chief hands it to the reviewer, which checks it on a different model.
- Step 4You mergeThe change and the review arrive side by side. What merges is your call.
What the team does, and what waits for you.
Waits for you
- Each command and file change while the bots are on Ask, the default
- Merging or pushing, when your brief says merge nothing
Start from a ready-made team.
From Murage’s library. Import one, look inside, and give it your brief.
- Ship Software SafelyGet a scoped implementation plan, coordinated engineering work, and a release-readiness review backed by verification evidence.4 bots
- Dev ShopTurn requests into technical plans, review quality and operational risks, and organize implementation handoffs and release priorities.4 bots
Skills the ready-made teams load.
What the library teams above bring with them. Skill Guard scans each one.
- AI Agent BuilderAI agent architecture covering agent patterns (ReAct, Plan-and-Execute, LATS), tool design, memory systems (short-term, long-term and episodic).
- AI Agent PatternsGuides expert-level ai agent patterns implementation: ai-ml and architecture decision frameworks and production-ready patterns.
- Monorepo ArchitectAdvanced monorepo architecture covering Nx, Turborepo, and Lerna tooling, workspace management patterns and build caching strategies.
- Scalability ArchitectScalability patterns expert covering horizontal vs vertical scaling, database sharding, read replicas, caching layers and CDN architecture.
- Search System ArchitectSearch infrastructure design expert covering full-text search architecture and Elasticsearch/OpenSearch cluster design.
- Operating rhythmIf they ask for a tool stack first, push back: rhythm before tools.
What you need.
What it takes to run this job. The Murage app is free; your AI provider bills you as usual.
- Claude Code and the Codex CLI, each signed in with your own plan. Murage opens the one-line install command for you
- Your project folder on this computer
- Murage open while the bots work
Rather have it set up for you?
We build your first engine review team with you, connect your tools and write your House Rules. Tell us what you need and we’ll quote before anything is booked.
Questions about engine review.
Why use two engines?
Different models miss different things. A reviewer on another engine reads the change fresh instead of checking its own work.
Can I use Hermes or another engine?
Yes. Each bot runs on the engine you choose for it: Claude Code, Codex, Hermes and more. Change a bot’s engine later and its job, notes, skills and routines stay put.
Does it cost extra?
The app is free. Each engine runs on the plan you already pay for.
Engine review: hand it to a team tonight.
Give your first job to Murage.
Download the free app, connect the AI you already pay for, and tell your Chief of Staff what needs doing. Plan on about ten minutes from install to a working team.
No account needed · Runs on the AI plan you already pay for