Your triage assistant, not your decision-maker

Stop reviewing
junk PRs.

PR Triage evaluates pull requests against the linked issue, repo patterns, and implementation substance. You get a structured triage decision in seconds, not a wall of AI text.

The problem is real

💥

Spam PRs waste your time

Bounty issues attract 10-25 PRs within hours. Most are cosmetic garbage dressed up as fixes. Reviewing each one takes 30+ minutes.

🗃

Good PRs get buried

When 80% of submissions are noise, quality contributions get delayed or lost. Contributors stop contributing. The project suffers.

Manual triage doesn't scale

Reading every diff across dozens of PRs? The review overhead exceeds the value of the contributions. Some maintainers just close everything.

How it works

1

Paste a PR URL

One input field. Paste any public GitHub PR URL. We fetch the diff, linked issue, and repo context automatically.

2

AI evaluates 6 dimensions

Issue fit, implementation substance, pattern alignment, scope match, test signal, and risk flags. Each one scored with evidence.

3

You get a triage decision

Structured score, confidence level, recommendation, evidence, and what to verify. Understand the PR in seconds, not minutes.

What you'll see

87/100
PrioritizeHigh PriorityHigh Confidence

PR directly addresses issue #342 by implementing retry logic in src/http/client.ts. Adds three test cases in tests/http/retry.test.ts covering timeout, 5xx, and connection reset scenarios. Changes are scoped appropriately to the networking module.

Issue FitStrong
SubstanceStrong
Pattern AlignmentModerate
Scope MatchStrong
Test SignalStrong
Risk FlagsStrong

Probabilistic assessment based on available context. Maintainers make the final call.

Six dimensions of review-worthiness

Issue Resolution Fit(30%)

Does the code actually address the linked issue? Compares the diff against what was requested.

Implementation Substance(25%)

Are the changes meaningful or cosmetic? Real logic changes vs. renames and formatting.

Pattern Alignment(15%)

Does the code follow your repo's conventions? Style, structure, error handling patterns.

Scope / Complexity Match(15%)

Is the change proportional to the problem? Flags both trivially small and suspiciously large PRs.

Test Signal(10%)

Were relevant tests added or modified? Do they actually verify the claimed fix?

Risk Flags(5%)

Red flags like unrelated file changes, suspicious patterns, dressed-up triviality.

Simple pricing

Start free. Upgrade when you need automation.

Free

$0

forever

  • ✓ Bring your own API key
  • ✓ Quick Scan analysis
  • ✓ 3 analyses per day
  • ✓ 7-day result history
  • ✓ Manual PR evaluation
Recommended

Pro

$19/mo

per maintainer

  • ✓ Everything in Free
  • Unlimited analyses
  • Deep Analysis mode
  • ✓ Permanent history
  • ✓ Shareable result links
  • ✓ Connect 5 repositories
  • ✓ Automatic PR ingestion
  • ✓ Export results (JSON/CSV)
  • ✓ Triage digest emails

Team

$49/mo

up to 10 members

  • ✓ Everything in Pro
  • ✓ 10 team members
  • ✓ Unlimited repositories
  • ✓ Team triage dashboard
  • ✓ API access for CI/CD
  • ✓ Custom triage rules
  • ✓ Priority support

BYOK: ~$0.002 per Quick Scan, ~$0.01 per Deep Analysis. You bring your own key. That's 500 quick scans for a dollar.

Ready to triage smarter?

Sign in with GitHub and analyze your first PR in under a minute.