Pull Request Explorer

Exploring 8 PRs for Anthropic Cookbook. Want this for your team? Start Free Trial

Title	Author	Size	AI	Cycle Time	Review	Merged
Rewrite the crop-tool cookbook around a measured zoom tool	cj-ant	XL	No	19.4h	0.1h	Jul 23, 2026
LLM analysis pending This PR has not been analyzed yet.
feat(managed_agents): add subagent live-streaming cookbook	cj-ant	L	No	1.3h	0.1h	Jul 22, 2026
LLM analysis pending This PR has not been analyzed yet.
Fix metaprompt notebook: remove assistant prefill (400 on current models)	nikblanchet	L	No	84.8h	0.1h	Jul 13, 2026
Summary Bug Fix Remove assistant prefill from metaprompt notebook Fixes a critical 400 error in the metaprompt notebook that prevented users from generating prompts, restoring functionality for all users. This ensures the tutorial works for all customers and eliminates a regression introduced by a model upgrade. Health Assessment AI Details Confidence: 0.95 Tech Stack
CI: fail the notebook-tests job when a changed notebook fails to execute	nikblanchet	S	No	10.4h	0.0h	Jul 10, 2026
Summary CI/CD Fail notebook-tests job on execution failures Ensures CI accurately flags notebook execution failures, preventing broken notebooks from passing tests. Health Assessment Small Low Low • Quick approval and merge indicate low complexity and minimal risk. AI Details Tech Stack Languages: Yaml Frameworks: Github-Actions
Add coordinator-pattern cost cookbook: big models plan, small models execute	bradabrams	XL	AI	94.2h	93.4h	Jul 06, 2026
Summary Feature Add cost cookbook for coordinator pattern Provides a cost‑effective multi‑agent workflow, demonstrating 2.5x cheaper and 3x faster execution, enabling teams to optimize resource usage. Health Assessment X-Large High High • PR required significant rework after initial review, with multiple config fixes and a large notebook addition, indicating complex changes and extended review cycle. AI Details Usage: Authored by AI Category: Code AI Tools: Claude Confidence: 0.95 Tech Stack Languages: Python, Yaml
fix(rag,classification): update rate-limit tier references to usage-tier naming	mattmccarley-ant	XS	No	0.1h	0.0h	Jul 02, 2026
Summary Docs Update rate-limit tier references to usage tiers Aligns documentation with current rate‑limit terminology, ensuring clarity for developers using RAG and classification guides. Health Assessment Small Low Low • Rapid approval with minimal changes indicates low complexity and risk. AI Details Confidence: 0.10 Tech Stack
Add roadtrip_planner managed agents cookbook	cj-ant	XL	No	0.4h	0.4h	Jun 30, 2026
Summary Feature Add roadtrip planner managed agent cookbook Provides a ready‑to‑run Next.js chat example that leverages managed agent sessions, enabling event streaming and multi‑agent coordination for rapid application prototyping. This cookbook accelerates development of agent‑powered chat experiences for developers. Health Assessment Large Low Low • The PR was merged within 0.4 hours of opening, with a single approving review and no rework, indicating high confidence in the cookbook’s quality and relevance. AI Details Tech Stack Languages: Typescript Frameworks: Nextjs, React
Reproduce Claude's agentic search benchmark scores in the Messages API	mengtingli-ant	XL	No	0.8h	0.1h	Jun 30, 2026
Summary Feature Reproduce Claude Agentic Search Benchmarks Adds a Messages API harness to reproduce Claude's DeepSearchQA and BrowseComp benchmark scores, enabling developers to validate agentic search performance and understand configuration impacts. Health Assessment X-Large Medium High • The PR was merged quickly with minimal rework, but the large code addition and multiple review rounds suggest a complex change that required careful validation. AI Details Confidence: 0.95 Tech Stack Languages: Python, Yaml

Get this analytics stack for your team

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