# AI use cases for Juror (Juror-AI/juror)

Repo: [Juror-AI/juror](https://github.com/Juror-AI/juror) | Use cases: 3 | Page: https://dedupedai.com/use-cases/juror-ai-juror/

Juror is a GitHub Action and CLI that has several frontier models review each pull request in parallel, merges duplicate findings, and posts one review with a merge score and cost receipt.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Add multi-model AI review to every pull request: https://dedupedai.com/use-cases/juror-ai-juror/multi-model-pr-review-in-github-actions/
- Review a single pull request from your terminal: https://dedupedai.com/use-cases/juror-ai-juror/review-one-pr-from-terminal/
- Check hosted review findings from an MCP client: https://dedupedai.com/use-cases/juror-ai-juror/check-hosted-reviews-from-chatgpt-via-mcp/

## How to add multi-model AI review to every pull request with Juror

Repo: [Juror-AI/juror](https://github.com/Juror-AI/juror) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README

Install the Juror workflow so each PR gets parallel reviews from several models merged into one report.

**Who it is for:** A team lead who wants AI code review without paying for a hosted review seat per developer.

**The problem:** A single AI reviewer is one opinion and can be confidently wrong.

**The outcome:** A posted PR review with merged findings, a merge score and a cost receipt.

### Steps

1. Export your key with `export JUROR_OPENAI_API_KEY=…`
2. Run `npx juror-ai init --set-secrets` in the repository
3. Or preview changes first with `npx juror-ai init --dry-run`
4. Open a pull request and read the posted review

### Prompt to paste into your agent

```text
Set up Juror-AI/juror in <repo> with `npx juror-ai init --dry-run`, then `npx juror-ai init --set-secrets`. Open a test PR and confirm the review includes a merge score and a cost receipt.
```

### Caveat

The sample workflow only reviews PRs from the same repository, so fork contributions are not reviewed automatically.

Risk, credentials: Provider keys are stored as GitHub secrets; the workflow skips fork PRs so the keys are not exposed to outside code.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/juror-ai-juror/multi-model-pr-review-in-github-actions/

## How to review a single pull request from your terminal with Juror

Repo: [Juror-AI/juror](https://github.com/Juror-AI/juror) | Difficulty: Intermediate | Time: 15 minutes | Basis: From the README

Run Juror locally on any PR and print the merged review, optionally posting it back.

**Who it is for:** A maintainer who wants a second opinion on a risky PR before approving it.

**The problem:** Setting up CI just to try a reviewer is overkill.

**The outcome:** A merged multi-model review printed to the terminal, or posted to the PR.

### Steps

1. Export provider keys in your shell
2. Run `npx juror-ai review --pr 1234 --repo owner/name`
3. Read the merged findings
4. Post it with `npx juror-ai review --pr 1234 --repo owner/name --post`

### Prompt to paste into your agent

```text
Run `npx juror-ai review --pr <number> --repo <owner/name>` and summarize the top findings. Confirm duplicate findings from different models were merged into one entry.
```

### Caveat

Cost depends on how many models you configure and the size of the diff.

Risk, credentials: The review uses your provider keys from the shell environment and each run is billed to them.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/juror-ai-juror/review-one-pr-from-terminal/

## How to check hosted review findings from an MCP client with Juror

Repo: [Juror-AI/juror](https://github.com/Juror-AI/juror) | Difficulty: Beginner | Time: 15 minutes | Basis: From the README

Connect the Juror Cloud MCP server to inspect workspace findings and rerun hosted reviews after confirmation.

**Who it is for:** An engineering manager who wants review summaries inside ChatGPT or Codex.

**The problem:** Switching to a dashboard to see review status breaks focus.

**The outcome:** Concise review summaries per workspace inside your MCP client.

### Steps

1. Add https://app.juror.dev/mcp as a server in your MCP client
2. Sign in with your Juror Cloud account
3. List your workspaces first
4. Inspect review summaries or ask it to rerun a hosted review

### Prompt to paste into your agent

```text
Through the Juror Cloud MCP server, list my workspaces and summarize open findings for <workspace>. Confirm no raw diffs were returned.
```

### Caveat

The MCP integration only covers repositories connected to Juror Cloud and never reads local repositories.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/juror-ai-juror/check-hosted-reviews-from-chatgpt-via-mcp/