# 技术写作

技术写作 类别下的全部知识中心文章。

URL: https://dylanengelbrecht.dev/zh/insights/category-authorship.html

- [为何选择 Mirror 做 Unity 多人联机](https://dylanengelbrecht.dev/zh/insights/mirror-multiplayer-unity.html): Mirror is a pragmatic default when Unity teams need a complete networking architecture.
- [渐进式披露式学习：地图，而非死记硬背](https://dylanengelbrecht.dev/zh/insights/progressive-disclosure-learning-with-agents.html): Orchestrate agents across a navigable map — learn foundations deeply, index everything else, verify with the environment.
- [Cursor 入门：自动化整个循环](https://dylanengelbrecht.dev/zh/insights/getting-started-with-cursor.html): Treat Cursor as an automation surface — rules, tests, MCP, and agents run the boring parts so you keep one thread for judgment.
- [面向 AI 编码代理的 AGENTS.md 标准](https://dylanengelbrecht.dev/zh/insights/agents-md-standard.html): One open Markdown file per repo — closest AGENTS.md wins; user prompts override.
- [为 AI 代理组织知识](https://dylanengelbrecht.dev/zh/insights/organizing-knowledge-for-ai-agents.html): Pair a public site with a private brain — progressive disclosure, lifespans, verification metadata.
- [用 search 与 execute 构建 MCP 服务器](https://dylanengelbrecht.dev/zh/insights/mcp-search-execute-codemode.html): Two MCP tools — search and execute — beat hundreds of per-operation tool definitions.
- [强化学习提升代理模型中的工具调用能力](https://dylanengelbrecht.dev/zh/insights/rl-tool-calling-composer.html): RL specializes models for the tool loop — Composer is the public case study.
- [Keep the thread: why plan–critique–build pipelines fail modern coding agents](https://dylanengelbrecht.dev/zh/insights/single-agent-coding-loops.html): One agent, one thread, small units — verify with tests, not lossy handoffs.