CLUSTER CODE-GITHUB-PULL-3 · 16 ENTRIES · EARLIEST ENTRY 2025-12-28
AI GitHub PR review tools
These products form a cluster of AI-powered code review tools focused on GitHub pull requests, combining static analysis with AI-generated, risk-categorized feedback. Dated entries before the index start are sparse: one in February 2025, a gap, then single entries from November 2025 through January 2026 and another in March 2026. ReviewScope shows the strongest traction signal with 25 stars, while patchcourt and pr-scout each show 1 star.
Entry timeline
ENTRY TIMELINE
Each bar is the month a member carries, which is a creation date where the source published one and a crawl date otherwise. Nothing in the index was found before 11 August 2026, and coverage broadened sharply on 14–15 August, so a bar from then on reports how much we found that month, not how much was built; an earlier bar reports how far back a member's own source dated it — not how long we have been watching.
Members by activity
Activity signals, not revenue.
Build origin · 1 ai-built · 15 unlabelled
Operator · 7 hobby · 1 indie · 1 commercial · 1 open source · 6 unlabelled
Two labels per entry, never collapsed into one: how it was built, and who stands behind it. Counted over the 16 listed below. Unlabelled means nothing has decided yet — not that the answer is small.
- 01patchcourt1★23 days—hobbyAI code review tool that analyzes GitHub pull requests, ties each claim to evidence, and returns a merge verdict.
- 02pr-scout1★16 days—Self-hosted pull-request triage tool that scans all open pull requests in a GitHub repository, scores them with AI, and presents a take-consider-skip board with reasons.
- 03ReviewScope25★6 mostaleliveopen sourceTool that reviews GitHub pull requests using static code analysis and AI models to surface low-noise, actionable issues.
- 04difflens0★26 days—hobbyHosted platform that reviews GitHub pull requests by combining static analysis with AI feedback verified against the exact code diff.
- 05CodeWarden0★16 days—hobbyCode review agent that watches GitHub pull requests for security vulnerabilities and standards drift, ranks issues by severity, and explains fixes.
- 06opensource-navigator0★24 days—indieTool that analyzes open-source repositories against a developer's skills and recommends contribution paths with relevant issues and code files.
- 07ForgeAI0★5 days—hobbyAutonomous software engineering tool that resolves GitHub issues by creating tested pull requests in a sandbox, repairing failures, and requiring human approval before merging.
- 08Ratchet0★10 daysliveWeb platform that detects and repairs code regressions from GitHub pull requests using automated analysis.
- 09gitanalyzer2★2 molivehobbyWeb tool that analyzes GitHub commit history to score message quality, surface writing patterns, and generate AI-assisted commit message suggestions.
- 10issuepilot0★1 mo—AI tool that triages GitHub issues by detecting duplicates, scoring priority, and suggesting actions.
- 11code_review_agent0★1 mo—hobbyAI code review workbench that analyzes Git diffs, pull requests, and merge requests, tracks findings, and exports Markdown, JSON, or SARIF reports.
- 12pr-review-ai6★9 mostalelivehobbyWeb tool that analyzes public GitHub pull request diffs with AI and generates structured, risk-categorized code review feedback.
- 13github_analyser6★1 yrabandonedliveTool that analyzes GitHub repositories to provide insights into code quality and structure.
- 14lgtm-feedback0★3 molivecommercialHosted platform that reviews GitHub pull requests using specialist AI agents, available as a web dashboard and command-line tool.
- 15Ultimate Code Metrics - AI-Powered GitHub Testing Intelligence0★unknownliveai-builtAI tool that analyzes test quality and recommends pull request approvals for GitHub projects.
- 16Diffwise — AI Code Review Tool for GitHub | 40+ Specialist Agents0★unknownliveAI code review tool that inspects GitHub pull requests in parallel with specialist agents, tracks regressions, and suggests fixes.
Ranked by traction, highest first, and paged 20 at a time — this is a window on the cluster, never the whole of it. Ties are broken by a stable id so a product cannot appear on two pages or fall between them. Browse every cluster.