OpenReview

Vercel Labs' self-hosted GitHub review bot: mention @openreview for Claude-powered inline suggestions; dormant beta since March 2026.

Visit Website [↗]

[ Facts ]

Category: AI PR Review
Open source: No
Pricing: Free to run; you pay your own Anthropic API and Vercel hosting. Beta with no commits since March 2026 (~1.5K stars, not archived)
Self-hosted: Yes — full stack — Self-hosted only: deploy the Next.js app to Vercel with your own GitHub App credentials and Anthropic key.
Platforms: github
Model control: BYOK Anthropic key on your own deployment; model hardcoded (Claude Sonnet 4.6), no provider choice without forking
Last verified: 2026-08-11

[ Against the 9 Standards ]

Based on public documentation as of 2026-08-11. ✓ documented · ~ partial · ✗ not offered · ? unknown. Methodology on the about page.

~ Multi-dimensional Context

Explores the codebase and runs project tooling during review; single repo, no persistent index.

~ Rule-Centric & Default Quiet

Reviews only when mentioned, quiet by design; extensible skills in .agents/skills, no severity config.

✗ Dual-Workflow: Local vs. PR

GitHub App reviewing PRs only; no IDE, CLI, or pre-commit surface.

✗ Business Logic Validation

No issue tracker or MCP integration documented; reviews the PR in isolation.

✗ Continuous Learning

Emoji reactions gate applying fixes per PR; no persistent learning from feedback.

~ Sandbox Validation

Runs project tooling (lint/format) as part of review; no documented test execution or preview environments.

~ Economic Transparency

Runs on your infra with your Anthropic key at zero markup; model hardcoded, no model independence.

✓ Actionability

Inline GitHub suggestion blocks with one-click apply; commits simple fixes directly to the PR branch.

✗ Measurable ROI

No dashboards or metrics of any kind.

[ In Our Coverage ]

Open Source AI Code Review: The Real Options (2026)

Open source AI code review tools compared: Kodus (AGPL), PR-Agent (MIT), and more — real licenses, BYOK costs, and how they stack up against closed SaaS.

Evaluating OpenReview?

Run it through the two-week trial protocol before you commit.

Evaluation Guide [↗]