OhMyBug

Finds 6× more real bugs than Cursor on ReviewBench.

The bug hunter for your coding agent. Catches the bugs your agent missed, then your own agent confirms and fixes every real one.

#1 on open ReviewBench ↗ as of November 11, 2026 The top-ranked bug‑finding reviewer: OhMyBug, for bug recall, ahead of Copilot and Devin
source on GitHub
Avoided: $1,000+prod bug
from $1per review · $4 fast
$0if clean · 1st free
Live · updated hourly

69,179 bugs caught this week

9,389 today · 6,355 reviews this week · 154 species in the Bugarium

Browser
install · two commands
claude plugin marketplace add OhMyX-AI/ohmybug-plugin
claude plugin install bughunter@ohmybug
Start free review

GitHub sign-in: your name only, no repo access.

Stop reviewing in circles. Close the loop.

Your agent reviews itself

  1. A long list: real bugs mixed with style notes and guesses.
  2. You check each claim against the code by hand.
  3. You re-run it, and a fresh list arrives.

Back to step 1.

Your agent + OhMyBug

  1. 1Your agent writes the diffas it does now
  2. 2OhMyBug hunts it in the cloudone deduplicated list; a second check drops weak claims
  3. 3Your agent verifies and fixesthe real bugs, in your repo
  4. Clean mergea hunt that finds nothing costs $0

Then the next diff, same loop.

CI billing · What leaves your machine · Privacy

What a hunt looks like

Illustrative session · sped up
jake — ✳ Claude Code — ~/acme/checkout-api — 120×36

Pick a speed. $0 if clean.

How to pick a mode ↗

Measured on OMBench: 85 held-out changes with 742 known bugs, November 2026. Typical answer is the median time on our side. Hard ones are 40 bugs that earlier reviews missed. Lite, snap and fast are from later runs.

Bug recall on ReviewBench

Recall · #1 on ReviewBench as of November 11, 2026

We replayed 171 real bugs at the commit that introduced them

Score 138 of 171 80.7%

124 caught · 28 partial, each worth half · 19 missed

  • django/django 59 fixes from 2026
  • postgres/postgres 98 fixes from 2026
  • public CVEs & OSS bugs 14

How the replay was run and scored

Sampled catches

  1. laravel/frameworkAuth token guard accepts array inputlived ~10 yearsCaught · deep hunt
  2. postgres/postgresHash bucket split leaks a buffer pinlived ~10 yearsCaught
  3. django/djangoHTML truncator denial of serviceCVE-2026-1285lived ~2 yearsCaught
  4. agno-agi/agnoCode execution via eval() on model outputCVE-2026-35002lived ~7 monthsCaught
  5. getsentry/sentry-javascriptAuth middleware bypass via the tunnel-route prefixGHSA-mchm-gjx6-7h55lived ~14 monthsCaught
  6. tokio-rs/tokioTimer data racelived 25 daysCaught

Bug species it catches

Browse Bugarium ↗

How it works

Technical note ↗
01

Diff + context

Your agent submits the change; the hunt runs in an isolated cloud workspace.

02

Hunt + 2nd pass

Drops false alarms ($0).

03

Your agent verifies

Billed from $1 per review if REAL (1 or 10 bugs) · $0 if clean.

Lite review · $1Snap review · $2Fast review · $4 · up to 15 minutesDeep review · $12 · up to 2 hours · $24 on a wide change · 1st free

Give your agent the reviewer ranked
#1 on ReviewBench. as of November 11, 2026

No card · first bug-finding review free$0 if cleanSAME HUNT IN CODEX

Flat per review, never per bug — from $1 (lite), $2 (snap), $4 (fast) or $12 (deep). Clean and minor-only runs cost $0; false alarms you reject cost nothing (how billing works).

Start free review →