Rejudge

RU

Two heads are better than one.
Several models and a judge dig deeper.
Independent second opinion for your agent.

Why several models

There are four ways to check code an agent wrote. Each one catches more than the last.

I. Same session, same model.
The agent rereads its own code. It already decided this code is correct, and reading it again does not change that decision.
II. New session, same model.
The context is clean, the model is not. It has the same training and the same habits, so it accepts the mistakes it would have written itself.
III. A different model, its own session.
A different model does find real bugs. But you now have two opinions, and when they disagree, you are the one who decides which is right.
IV. Rejudge: three models and a judge.
All three models get the same question at once. Each one works in an isolated context and makes its own tool calls, so none of them sees what the others do. The judge reads all three reports, asks follow-up questions where they disagree, and writes one answer. When all three report the same problem, three independent checks agree on it. When they split, the judge resolves it instead of you.

How it works

Your agentasks a questionModel Xwrites its findingsModel Ywrites its findingsModel Zwrites its findingsJudgecompares the findingsAnswerbased on all findings+ Run IDto resumeYour agentasks a questionModel Xwrites its findingsModel Ywrites its findingsModel Zwrites its findingsJudgecompares the findingsAnswerbased on all findings+ Run IDto resume
  1. Each model gets the same request and works it independently of the others.
  2. The judge checks their answers and, where needed, asks more questions or sends the work back.
  3. The judge returns one finished answer, put together from what the models found.

Quick start

01 Install Rejudge and its skills

npm install -g rejudge; npx skills add syabro/rejudge -g -y

02 Using Pi? Add the extension

pi install "$(npm root -g)/rejudge"

Inside Pi you watch the run live and see which model is at which stage. Agents get more control over the tool than over the CLI.

03 Connect a provider

Already using Pi? Nothing to do here. Rejudge takes the providers you set up in Pi.

Rejudge runs on Pi and reads Pi's provider settings, so any key Pi accepts works here: ANTHROPIC_API_KEYOPENAI_API_KEYGEMINI_API_KEYOPENROUTER_API_KEYOPENCODE_API_KEY and more — the full list is in the Pi docs.

I personally use OpenCode Go (referral link: $5 for you, $5 for me) because it offers an excellent mix of models for $10 a month.

Using a subscription instead of API keys?

For subscription logins Rejudge still goes through Pi. If Pi is not authorized yet, run:

npx -y @earendil-works/pi-coding-agent

Inside Pi, run /login and finish sign-in with your subscription provider.

04 Pick your models

Create ~/.config/rejudge/config.json once.

{
  "reviewers": [
    "opencode-go/deepseek-v4-pro@high",
    "opencode-go/mimo-v2.5-pro@high",
    "opencode-go/minimax-m3@high"
  ],
  "judge": "opencode-go/glm-5.1@high"
}

Two reviewers minimum. Every model needs a reasoning level, and a higher level means a longer and more thorough review: minimal, low, medium, high, xhigh.

05 Ask

ask rejudge to analyze src/payments and list the potential security holes
/rejudge compare docs/adr/012-event-bus.md with the code and list the gaps
/rejudge-diff
rejudge "go through migrations/0042_add_index.sql and list the risks"
git diff | rejudge "review this change"
rejudge --resume <run-id> "what about the rollback path?"