Anthropic Cookbook Engineering Benchmarks

Anthropic · GitHub · Updated June 3, 2026

Anthropic Cookbook delivery benchmarks are based on 15 merged pull requests in the last 30 days (81 all-time), with 8.2 hours median cycle time, 0.1 hours median review time, and 7 active contributors. AI-related signals appear on 26.7% of recent work. PR volume is up 114% compared to the prior period.

Median Cycle Time
8.2h
PR open to merge
Median Review Time
0.1h
first review to approval
AI-Related Signals
26.7%
directional signal across recent PRs
Best Signal Growing Cadence
+114%
PR volume up 114% vs prior period

Repo vs Anthropic Average

+4.7pp
AI Signals
+5.4h
Cycle Time
0.0h
Review Time

Delivery Trend and AI Signals

PR Backlog (Humans)

Excludes AI agents and automation accounts.

PR Backlog (AI Agents)

AI coding agents only.

AI Signals vs Cycle Time

Correlation

moderate positive
r = 0.49 (10 weeks)

Cycle Time by AI-Related Signals

4.8h
with signals
81.9h
baseline
-94.2%
difference

PR Breakdown

2 with signals / 15 total PRs

Top AI Tools

claude_code 1
claude 1

PR Type Breakdown

feature 9
refactor 2
ci 2
chore 1
bugfix 1

Methodology

Metrics are computed from merged pull request data synced daily from GitHub. AI-related metrics are directional signals inferred from pull request activity patterns and heuristics. Summary metrics use a 30-day rolling window; trends use a 90-day window. Last updated 16 hours, 29 minutes ago.

Updated daily from public GitHub pull requests.

See Your Team's Benchmarks

Get this analytics stack for your team

Connect GitHub and see cycle time, review bottlenecks, PR flow, and trend changes in minutes.

Connect Repos