ProPR
Log in Get ProPR

Quickstart: setup to your first pull request.

Two commands bring up the stack. Then add a repository, label an issue AI, read the pull request ProPR opens, and comment on it to get the follow-up commit.

propr setup Guided
  1. Verify hostDocker + Node.js
  2. Authorize agentsexisting credentials
  3. Connect GitHubrelay or your App
  4. Start stackpublished images
propr setup checks each boundary before the stack starts.

Setup overview

Check the requirements

A Linux amd64 host with Docker available to your user, Node.js 22+ and npm, a GitHub account, and an account with one supported agent provider. Allow 2 vCPU, 4 GB RAM and 20 GB of disk to evaluate, and 8 GB RAM for daily use. ProPR needs outbound HTTPS and WebSocket access and no inbound public port. Full system requirements.

Install the CLI and run propr setup

Install propr-cli and run the guided setup from an empty directory.

npm install -g propr-cli
propr setup

propr setup checks the host, signs in your agents, connects GitHub through ProPR Connect, and starts the published images on your Docker daemon. GitHub and provider sign-in happen in your browser. It finishes by offering to add a first repository and opening the Web UI at http://localhost:5173.

  • Server or teamPublic URL and TLS behind a reverse proxy: server setup.
  • Hardened VPSLocked-down host before the install: secure VPS deployment.
  • Your own GitHub AppRegister a private App with propr github-app create: GitHub authentication.
  • macOS on Apple SiliconDocker Desktop runs the published amd64 images: local setup.
  • Your coding agent installs itPaste the installation prompt into your agent chat and authorize GitHub and your provider when it pauses.
propr setup, start to finish.The guided CLI flow from install to the dashboard.

Sign in and add a repository

Open http://localhost:5173, sign in with GitHub, add the repository and enable an agent. Your GitHub user is already the instance administrator.

propr repo add owner/repo

The same works from the terminal once propr remote and propr login point the CLI at the stack. Coding agents shows each agent with its health and default model.

Coding agents: Claude Code, Codex, Mistral Vibe, Antigravity and OpenCode, each with its health and default model
Coding agents after setup: each one with its login state and default model.

Label an issue AI, or start from a plan

Write a normal GitHub issue and add the AI label. A model label such as llm-claude-opus55 picks the agent and model; several model labels on one issue produce one pull request per model. To shape the work first, describe the change in Planner Studio, refine the plan in chat and create the issues from it. From the CLI, create the plan, list its issues, then implement one of them:

propr plan create "Add dark mode" --wait
propr plan issues <draft-id>
propr issue implement <draft-id>/1 --wait

ProPR picks the issue up, tracks progress with a state label, and shows the task on the dashboard with its live log: each edit explained, each test run.

The execution trace of a live run in plain language: why the time-zone fix is needed, the new DST test cases, and all 14 checkout tests passing
The live log of the run: the fix, the new tests, and the test results.

Read the pull request

ProPR commits, pushes and opens a pull request linked to the issue. The agent writes code in an isolated Docker container and Git worktree; ProPR makes every branch, commit and push.

The pull request carries the diff and a run summary: status, duration, token cost and model. The task view keeps the full prompts, the live log and the per-call cost. Review the diff in GitHub the way you review any pull request. The longer version of this path, with what each step needs, is the guide to an AI coding agent that opens pull requests from issues.

Task header: repository and pull request, Implementing status, the run in progress, its model and runtime, and token consumption
The task header in the Web UI: the pull request, the run, its model, runtime and tokens.

Comment, and get the follow-up commit

Write what you want changed as a plain comment on the pull request or on a diff line.

@maintainercommented on this pull request

Please tighten the error handling in the auth callback.

ProPR pushes one more commit to the same branch and replies with links to the task, the commit and an undo. /review posts a scored critique, and /fix applies its findings. You merge in GitHub.

A follow-up run on the pull request: read the PR comment about the phone layout, move the summary below the form, run the checkout tests
The follow-up run: read the comment, make the change, run the tests.

Rather skip the server?

ProPR Hosted runs a dedicated instance for you, and you start at step three.

See Hosted plans