Claude Code v2.1.283 ships prompt-audit doctor command and model blocking controls
ashwin-ant · ghdev · 2026-09-26
Claude Code v2.1.283 adds a /doctor prompt-audit command that flags prompting patterns in CLAUDE.md, skills, agents and commands written for older models. Enterprise settings gain availableModelsMatch and deniedModels for precise model gating, the gateway gets a prompt-id hint header plus a load-test mode and Bedrock Mantle provider, and multiple MCP bugs are fixed — including lost background progress notifications, stdio servers left running, and remote servers becoming unusable after a 404.
More from coding & agent
- Hamel Husain, BEBischof and isaac_flath rank AI engineering techniques in a live tier list — HamelHusain · 2026-09-26
- In an AI-first world, your own understanding of the project is state to preserve — brandon_xyzw · 2026-09-26
- Railway launches free no-account VMs (59 min) and OpenCode cloud agents — jasonkneen · 2026-09-26
- 575 PRs later: how the Astra agent untangled a sync-SQLite bottleneck in OC — soumitrashukla9 · 2026-09-26
- From GoPro footage to 3D scenes: a two-week agent skill for full-home reconstruction — doodlestein · 2026-09-26
- Sentinel splits monolithic Three.js into scene-based workflows with HD PNG sequence export — PurzBeats · 2026-09-26