Claude 工程实践

25

25 篇文章

  1. 重新设计 Anthropic 的源代码管理(source control)

    Redesigning source control at Anthropic

  2. 我们如何跨产品约束 Claude

    How we contain Claude across products

  3. 关于近期 Claude Code 质量报告的更新

    An update on recent Claude Code quality reports

  4. 扩展 Managed Agents:将大脑与双手解耦

    Scaling Managed Agents: Decoupling the brain from the hands

  5. 我们如何构建 Claude Code auto mode:一种更安全地跳过权限的方式

    How we built Claude Code auto mode: a safer way to skip permissions

  6. 面向长时间运行的应用开发的框架设计

    Harness design for long-running application development

  7. Claude Opus 4.6 在 BrowseComp 表现中的评测意识(eval awareness)

    Eval awareness in Claude Opus 4.6’s BrowseComp performance

  8. 用一组并行 Claude 构建 C 编译器

    Building a C compiler with a team of parallel Claudes

  9. 量化智能体编程评测(eval)中的基础设施(infrastructure)噪声

    Quantifying infrastructure noise in agentic coding evals

  10. 设计抗 AI 的技术评测

    Designing AI-resistant technical evaluations

  11. 揭开 AI 智能体评测的面纱

    Demystifying evals for AI agents

  12. 面向长时间运行智能体(agent)的有效框架(harness)

    Effective harnesses for long-running agents

  13. 在 Claude 开发者平台上推出高级工具调用

    Introducing advanced tool use on the Claude Developer Platform

  14. 使用 MCP 进行代码执行:构建更高效的智能体

    Code execution with MCP: Building more efficient agents

  15. 超越权限提示:让 Claude Code 更安全、更自主

    Beyond permission prompts: making Claude Code more secure and autonomous

  16. 用 Agent Skills 为现实世界装备智能体

    Equipping agents for the real world with Agent Skills

  17. 为 AI 智能体(agent)进行有效的上下文工程(context engineering)

    Effective context engineering for AI agents

  18. 对最近三个问题的复盘(postmortem)

    A postmortem of three recent issues

  19. 为智能体编写有效的工具——借助智能体

    Writing effective tools for agents — with agents

  20. Desktop Extensions:为 Claude Desktop 提供一键安装 MCP 服务器

    Desktop Extensions: One-click MCP server installation for Claude Desktop

  21. 我们如何构建我们的多智能体研究系统

    How we built our multi-agent research system

  22. “think”工具:让 Claude 在复杂工具调用场景中停下来思考

    The "think" tool: Enabling Claude to stop and think in complex tool use situations

  23. 借助 Claude 3.5 Sonnet 在 SWE-bench Verified 上抬高标杆

    Raising the bar on SWE-bench Verified with Claude 3.5 Sonnet

  24. 构建高效的智能体

    Building effective agents

  25. 引入上下文检索(Contextual Retrieval)

    Introducing Contextual Retrieval

Anthropic Engineering