MCP 与 Skills

覆盖 2026-09-11 至 09-16 的 AI Agent 工程阅读日志，保留本期重点、工程意义、优先阅读位置与一手来源。

AI agent engineering readings for September 11 to 16, 2026, with key findings, engineering implications, suggested sections, and primary sources.

Skills Manager distributes a local skill library to Claude Code, Codex, Cursor, and Gemini CLI while managing version and configuration consistency.

介绍 Skills Manager 如何用本地 Skill Library 统一分发到 Claude Code、Codex、Cursor 和 Gemini CLI，并处理版本与配置一致性。

How the Google Cloud Developer Plugin packages Skills, MCP configuration, and rules, and what this means for distribution, versions, and permissions.

拆解 Google Cloud Developer Plugin 如何把 Skill、MCP 配置和规则按 Agent Plugins 规范打包，暴露跨 Coding Agent 的分发、版本与权限问题。

AI agent engineering readings for September 4 to 10, 2026, with key findings, engineering implications, suggested sections, and primary sources.

覆盖 2026-09-04 至 09-10 的 AI Agent 工程阅读日志，保留本期重点、工程意义、优先阅读位置与一手来源。

Munim Computer Use reads the accessibility tree and operates desktop controls by element ID, reducing reliance on fragile screenshot-coordinate automation.

介绍 Munim Computer Use 如何先读取可访问性树，再按元素 ID 操作桌面控件，减少依赖截图坐标的脆弱自动化。

Strands harness-sdk, Harness MCP Server, and HydraFusion show why agent engineering needs control over context, permissions, state, validation, and recovery.

从 Strands harness-sdk、Harness MCP Server 和 HydraFusion 看 Agent 工程为什么从模型调用转向上下文、权限、状态、验证与恢复控制。