Frequently Asked Questions

Does my code leave my infrastructure? Repository clones, plans, task records, logs, and credentials stay in your self-hosted stack. Two external systems see parts of your work: GitHub (branches, PRs, comments — where it already lives) and the model provider you route a task to (the prompt and code context for that task). The optional ProPR Connect relay forwards GitHub webhook events to your stack but never receives repository contents. See the Security Overview.

Do I have to use ProPR Connect? No. Connect is the zero-config path (shared GitHub App + token relay + delivery replay). You can register your own GitHub App instead and receive events by direct webhook or polling — no ProPR-hosted service involved. See GitHub Authentication.

What does ProPR cost? The software is free and open source (Apache 2.0). You pay your AI providers directly — existing subscriptions work, and ProPR never marks up tokens. ProPR Connect is free for up to 3 users per installation; more seats and the managed hosted-UI tunnel are part of the paid Plus tier.

Which agents and models can I use? Claude Code, Codex, Antigravity, OpenCode, and Mistral Vibe — on subscription-backed logins or API keys, switchable per task and per phase with llm-* labels or /switch. See Agents and Models.

How is this different from running the agent CLI myself? A terminal session needs you at the keyboard, approving commands and edits as the agent works. ProPR runs the same agents unattended: the isolated workspace is the permission boundary, and your approval moves to the pull request. A well-scoped task becomes delegation — label the issue, walk away, review the PR. Follow-up feedback is lighter too: screenshots paste straight into GitHub comments, and ProPR hands them to the agent.

Is the agent sandboxed? Each task runs in its own Docker container and Git worktree, and ProPR itself performs all Git and GitHub operations on the agent's behalf. Outbound network access is open by default; the optional restricted network mode starts agent containers with no network and routes them through a per-run allowlist proxy on the worker, with no privileged containers. See Execution Safety.

Can I run it offline or air-gapped? No. ProPR needs GitHub and your model providers to do its job. The documentation is bundled for offline reading (propr docs on), but the workflow itself is GitHub-centered.

Does it work with GitLab or Bitbucket? No — the whole loop is built on GitHub's APIs: pull requests, review comments, checks, and comment commands.

What happens when a run fails? The task record keeps the prompt, logs, and failure state; issues get AI-failed-* labels; and you retry, switch models, or revert from the PR conversation or dashboard. See Troubleshooting.

Who can make ProPR do work? A user whitelist gates the dashboard, the CLI, and GitHub-triggered work; bots are filtered; PR commands run only for allowed authors. See Who Can Trigger Work.

Where is my data stored, and how do I remove it? In the stack directory on your host (data/, logs/, repos/), the Redis volume, and repository clones and worktrees under /tmp/git-processor by default. The Teardown guide lists every artifact to remove.