PR Quorum

PR Quorum is a cloud-based GitHub App that reviews pull requests using configurable AI reviewers and OpenRouter. It runs specialist reviewers, removes duplicate and low-confidence findings, and posts a focused review to GitHub. Its default panel covers Correctness, Security, and Architecture; five more templates and custom reviewers are available. Reviewers can have separate models, focus lists, confidence thresholds, and posting behavior, with up to 20 in a panel. Repository settings in .ai-review.yml can specify triggers, skipped files or pull requests, thresholds, inline-comment limits, and reviewer options. Reviews run when pull requests are opened, updated, reopened, or marked ready for review; drafts are skipped by default. The service posts comments and status checks, but does not block merges or approve them automatically. The free plan includes 50 reviews per month across two repositories, with a three-reviewer panel and up to three inline comments. Pro is listed at $9.60/user/mo (annual). The REST Developer API is in preview and is not yet generally available.

PR Quorum provides customizable review feedback within a GitHub pull-request workflow, with free and paid plans. Its reviews are advisory, so they do not control whether a change is merged.

PR Quorum is presented for solo developers, agencies, and teams using GitHub pull requests. Its configurable reviewers and repository rules suit users who want review feedback tailored to their codebase.

PR Quorum is a configurable AI review service for GitHub pull requests. It suits solo developers, agencies, and teams that want several specialized perspectives without moving review work out of GitHub. Its main advantage is unusually granular reviewer and repository control; its main constraint is that every result remains advice rather than merge enforcement.

Overview

PR Quorum sends a pull request through specialist reviewers, removes duplicate and low-confidence findings, and publishes one focused review in GitHub. The default panel covers correctness, security, and architecture, while five additional templates and custom reviewers allow a panel tailored to a project. Each reviewer can use a separate model, focus list, confidence threshold, and posting behavior, with up to 20 reviewers in one panel.

Reviews start when a pull request is opened, updated, reopened, or marked ready for review; drafts are skipped by default. A repository-level .ai-review.yml file can define triggers, skipped files or pull requests, confidence thresholds, inline-comment limits, and reviewer settings. That makes policy reproducible beside the code, although the service still cannot stop a merge or approve one automatically.

Key features

The GitHub App routes model requests through OpenRouter, while pull-request webhooks queue review runs through Inngest. HMAC verification protects GitHub event handling, and the App requests only the privileges needed to post a review. PR Quorum says it does not train models on customer code or sell customer data, and sends zero-data-retention flags when a provider supports them. OpenRouter keys are encrypted at rest with AES-256-GCM and decrypted only in the review worker.

Usage events record model, token count, reviewer timing, failures, and estimated cost for the cost dashboard. That is useful when teams need to understand model spending, but the REST Developer API is still in preview and is not generally available. The cloud service supports security analysis, custom review rules, and fix suggestions, with GitHub as its code host.

Pricing

The Free plan costs nothing, billed forever. It includes 50 reviews per month, two repositories, a three-reviewer panel, and a maximum of three inline comments per review, with Community support. It is a sensible way for an individual to assess the workflow, but the repository, panel, and comment ceilings make it a poor fit for broader team coverage.

Pro costs $115.20 a year, billed yearly ($115.20/developer), and provides 150 pooled reviews per seat each month, unlimited repositories, your own OpenRouter key, a custom reviewer panel, per-path rules, reviewer modes, a cost dashboard, and 10 inline comments. Email support and the higher review allowance suit a developer or small team that needs control across many repositories.

Team costs $20 a year, billed yearly ($240/developer), with 300 pooled reviews per seat each month. It includes everything in Pro plus members and roles, an audit log, outbound webhooks, saved views, and priority support. Those governance features make it the more appropriate tier for an organized team, though the displayed annual amount is higher than Pro's despite the plan label's lower figure.

Max costs $40 a year, billed yearly ($480/developer), and raises pooled volume to 750 reviews per seat monthly. It adds 15 inline comments per review and the highest platform-paid review volume, with priority support. Enterprise uses custom pricing and adds SSO and SAML, self-host support, an SLA and security review, custom usage limits, and dedicated support. There is no free trial.

Platforms

PR Quorum is delivered as a cloud service through web and API access, centered on its GitHub App. The developer API is preview-only, so organizations requiring a stable, generally available API should wait or choose a more mature integration.

Who it's for

Choose PR Quorum when you want repository-local rules, distinct reviewer personas, configurable confidence thresholds, and visibility into model costs. Solo developers can start within the free cap; agencies and teams gain more value from pooled quotas, unlimited repositories, roles, audit history, and webhooks. It is not the right choice if automated approval or merge blocking is a requirement.

Pros and cons

• Pros: Up to 20 independently configured reviewers; duplicate and low-confidence findings are consolidated; repository policy can live in .ai-review.yml; security controls include HMAC verification and encrypted OpenRouter keys; paid plans expose usage and estimated cost.
• Cons: The free tier stops at two repositories, 50 monthly reviews, three reviewers, and three inline comments; paid quotas are pooled per seat and vary sharply by plan; the API is not generally available; reviews cannot approve or block merges.

Alternatives

See the AI Code Review Tools roundup for broader comparisons. Pick CodeRabbit instead if you need a free trial and a wider platform range, including extensions, desktop systems, and self-hosting. Ellipsis is another freemium API and web option, with an Individuals plan tied to a Claude Code or Codex subscription. Choose Gitar if GitHub or GitLab access, self-hosting, and an open-source plan matter. AI Code Reviewer is a free, self-hosted Linux option where you pay only your model provider. Codacy fits developers wanting IDE integrations and support for 38 languages. CodeAnt AI offers a 14-day trial and self-hosting. CodeKang provides a forever-free tier with five private repositories. Cubic offers free public-repository reviews with a 20-review monthly starter allowance.

Verdict

PR Quorum is best for GitHub users who need finely tuned, multi-perspective review feedback and repository-level policy, especially when cost tracking matters. Choose it for configuration depth and human-controlled merges; look elsewhere for generally available APIs, broader host support, or automation that can enforce merge decisions.