Claude writes 90% of Claude Code's code; merged PRs doubled while reverts fell 40%
PrajwalTomar_ · x · 2026-10-19
The Claude Code team reveals Claude now writes 90% of its code: from June to September, merged PRs more than doubled while the share reverted fell over 40%. Their published review process: run /code-review and /security-review before PRs or use a fresh-context subagent; move repeated review comments into REVIEW.md; tier review effort by risk (quick pass on routine changes, deep review where bugs cost money); never let a bot's approval replace a human for auth, billing, releases; use merge queues, feature flags, staged releases and one-step rollback; turn incidents and reverts into an eval set and rescore the reviewer when prompts, tools or models change.
More from coding & agent
- Websites Are Closing Their Doors to AI Agents: Yelp, Delta, JPMorgan Block Bots — steveplunkett · 2026-10-19
- 15 AI Agent Design Patterns, Distilled From Building 400+ Agents — MaryamMiradi · 2026-10-19
- Open-source Groki bot turns an ESP32 into a voice-controlled Grok desktop assistant — alexcovo_eth · 2026-10-19
- AI agent rebuilds black hole images from raw EHT data, matching 91%-98% of official results — fofrAI · 2026-10-19
- Grok bot doesn't code: it delegates tasks to Claude Code sessions on your Mac — alexcovo_eth · 2026-10-19
- Delegating Work to AI Agents by Email: Send a Mail, Get SEO Analysis Back — pramodk73 · 2026-10-19