Engineering (工程博客)
Anthropic 工程实践文章，涵盖 Agent 构建、工具使用、评估、上下文工程、MCP 等主题。
| 文章 | 标题 |
|---|---|
| AI-Resistant Evaluations | Designing AI-Resistant Technical Evaluations |
| Postmortem of Three Issues | A Postmortem of Three Recent Issues |
| Advanced Tool Use | Introducing Advanced Tool Use on the Claude Developer Platform |
| Building a C Compiler | Building a C Compiler with a Team of Parallel Claudes |
| Building Effective Agents | Building Effective Agents |
| Claude Code Best Practices | Best Practices for Claude Code |
| Claude Code Sandboxing | Beyond Permission Prompts: Making Claude Code More Secure and Autonomous |
| Think Tool | The 'think' Tool: Enabling Claude to Stop and Think in Complex Tool Use Situations |
| Code Execution with MCP | Code execution with MCP: Building more efficient agents |
| Contextual Retrieval | Introducing Contextual Retrieval |
| Demystifying Evals | Demystifying Evals for AI Agents |
| Desktop Extensions | Claude Desktop Extensions: One-click MCP Server Installation |
| Context Engineering | Effective Context Engineering for AI Agents |
| Long-Running Agent Harnesses | Effective Harnesses for Long-Running Agents |
| Eval Awareness BrowseComp | Eval Awareness in Claude Opus 4.6's BrowseComp Performance |
| Infrastructure Noise | Quantifying Infrastructure Noise in Agentic Coding Evals |
| Multi-Agent Research System | How We Built Our Multi-Agent Research System |
| SWE-bench Sonnet | Raising the bar on SWE-bench Verified with Claude 3.5 Sonnet |
| Writing Tools for Agents | Writing Effective Tools for Agents |