Event ·Official first-hand

Claude Code 发布 v2.1.283：新增模型管控设置与多项 MCP 修复

Anthropic 发布 Claude Code v2.1.283，新增 availableModelsMatch 与 deniedModels 两个托管设置，管理员可精确锁定可用模型版本并封禁指定模型。该版本还在网关提示头中加入 x-claude-code-prompt-id，便于 LLM 网关归类同一用户提示的请求。同时修复了 SDK 会话丢失 deferred tool call、MCP 进度通知被丢弃、stdio MCP 服务器会话结束后未关闭等多个问题，并新增 /doctor prompt-audit 命令检查为旧模型编写的提示词。

Current read

Claude Code 更新到 v2.1.283 了，管理员终于能精确控制团队能用哪些模型版本，还修了一堆 MCP 的坑，重度用户建议升级。

11 sources58° AI heatLast updated 2026/9/25 21:50:12

Evidence

11 sources

Primary sourceClaude Code: GitHub Releases

2026/9/25 21:50:12View source

Related source 1小互

2026/9/24 01:37:20View source

Related source 2claudedevs

2026/9/23 22:49:38View source

Related source 3arXiv: OpenAI

2026/9/24 02:00:20View source

Related source 4IT之家

2026/9/24 04:01:11View source

Related source 5岚叔

2026/9/25 19:38:52View source

Related source 6TestingCatalog

2026/9/26 00:09:11View source

Related source 7Thariq

2026/9/24 19:56:43View source

Related source 8宝玉

2026/9/26 02:52:12View source

Related source 9Artificial Analysis

2026/9/24 01:24:22View source

Related source 10Riley Goodside

2026/9/25 15:21:08View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story