Release v2.1.283 anthropics/claude-code

Release v2.1.283 · anthropics/claude-code

• ID: 698a7c41
• 原文链接: https://github.com/anthropics/claude-code/releases/tag/v2.1.283
• 作者: ashwin-ant
• 日期: 2026-09-25
• 分类: coding
• 来源类型: github
• 标签: claude-code, release-notes, gateway, mcp, enterprise
• 质量评分: 3/5
• 抓取时间: 2026-09-30T15:48:51Z

中文导读

Claude Code v2.1.283（9 月 25 日发布）是一版明显偏向托管与企业部署的更新：新增 availableModelsMatch 管理设置，设为 exact 时 availableModels 只放行点名的模型版本，新发布模型默认被挡在外面，另加 deniedModels 显式黑名单；网关提示头新增 x-claude-code-prompt-id，让 LLM gateway 能把服务同一 user prompt 的请求分组对账；OTEL 的 tool.output 事件可选带出 MCPWebFetch 与 WebSearch 输出；新增 /doctor prompt-audit 审计 CLAUDE.mdskillsagents 里写给旧模型的提示模式；还包含 gateway load_test_mode（构造并签名请求但不发往上游）Bedrock Mantle 上游，以及一批 MCP 修复：后台任务进度stdio 服务器残留进程stateless 服务器 404 后整会话不可用等

为什么值得关注

Claude Code 企业闸门版本：模型 pin 策略、prompt-id 对账与 MCP 稳定性修复，托管团队可直接执行。

关键信息

• 标题: Release v2.1.283 · anthropics/claude-code
• 作者: ashwin-ant
• 原文链接: https://github.com/anthropics/claude-code/releases/tag/v2.1.283
• 发布时间: 2026-09-25
• 来源平台: github
• 关联标签: claude-code, release-notes, gateway, mcp, enterprise

English Summary

Claude Code v2.1.283 (released Sep 25) leans toward managed and enterprise deployments: the new availableModelsMatch managed setting with "exact" restricts availableModels entries to the exact version named, so new releases stay blocked until listed, plus a deniedModels blocklist; gateway hint headers gain x-claude-code-prompt-id so LLM gateways can group requests serving one user prompt; OTEL tool.output span events can opt into MCP, WebFetch and WebSearch content; /doctor prompt-audit audits CLAUDE.md files, skills, agents and commands for prompting patterns written for older models; plus a gateway load_test_mode (requests built and signed but not sent upstream), a Bedrock Mantle upstream, and MCP fixes spanning background progress notifications, leftover stdio servers, and stateless-server 404 lockouts.

English Source Excerpt

What's changed

- Added x-claude-code-prompt-id to the gateway hint headers so LLM gateways can group the requests that serve one user prompt; opt in with CLAUDE_CODE_GATEWAY_HINT_HEADERS=1

- Added availableModelsMatch managed setting: with "exact", an availableModels entry allows only the model version it names, so new releases stay blocked until listed

- Added deniedModels managed setting to block specific models, even when availableModels allows them

- Added MCP tool, WebFetch and WebSearch outputs to the tool.output OpenTelemetry span event when OTEL_LOG_TOOL_CONTENT=1

- Added /doctor prompt-audit (also /checkup prompt-audit) to audit your CLAUDE.md files, skills, agents and commands for prompting patterns written for older models

- Added click-to-expand for truncated messages from your other sessions in fullscreen mode

- Added path to --plugin-dir load-failure entries in the stream-json system/init plugin_errors, naming the directory that did not load

- Added an opt-in load_test_mode block to the Claude apps gateway config: requests are built and signed but not sent upstream, and clients get a canned reply, so a deployment can be

Obsidian Notes

• 内容由 opencli web read 抓取原文页面生成。
• 中文导读与英文摘要基于当日抓取正文与 Obsidian 摘要笔记交叉核对；英文摘录为原文逐字片段。