Public PR review timing snapshot for onyx-dot-app/onyx

[xpinc17]

Hi Onyx maintainers,

I put together a small read-only analysis of recent public PR review timing for onyx-dot-app/onyx. This uses only public GitHub data, with no app install and nothing posted back to the repository.

The short version: most PRs appear to move very quickly, but at this volume there is still a visible aging tail, especially after approval.

For PRs created between 2026-03-10 and 2026-06-08:

• 2,407 PRs were analyzed
• 22 PRs waited more than 48h for a first review
• 63 approved PRs remained unmerged for more than 48h after their last approval
• Median first-review wait was about 0.1h
• Cumulative observed first-review wait was about 100.5 PR-days

So this does not look like a broad review-speed problem. It looks more like a high-throughput queue where most PRs move quickly, while a few aging review items can still slip out of view, particularly after approval.

A few longest observed first-review waits from the window:

• #10179 — 14.4 days
• #11515 — 7 days
• #11517 — 167.7h

One retrospective estimate from the same window: surfacing aging review items earlier would have flagged about 21% of the observable review-wait time, or roughly 20.7 cumulative PR-days. This is an estimate from public data, not a guarantee.

Method caveats:

• This is a historical, read-only estimate from public GitHub data.
• It may not reflect internal reviewer load, branch protection, prioritization, or project policy.
• Aggregate first-review wait may use merge/close time as a fallback when no first-review timestamp is visible.
• The longest-wait examples only include PRs with observed first-review timestamps.

I’m building ApprovalOps, a tool for surfacing aging review items earlier. I’m sharing this because the snapshot seemed potentially useful for a high-volume project like Onyx, and I’d be happy to share the methodology or run another read-only window if useful.

ADR Report: skipped / ADR not requested / n/a / n/a