anthropics / cwc-workshops

cwc-workshops Workshop materials. Not maintained and not accepting contributions. Materials from Anthropic-run Code with Claude workshops.

cwc-workshops 研讨会资料。本项目目前已停止维护，不再接受贡献。这些资料来自 Anthropic 举办的“Code with Claude”研讨会。

rightmodel/ — Picking the Right Model: use a Claude Code SKILL to audit an LLM eval suite and sweep it across models and inference parameters (extended thinking, effort) to find the best quality-per-dollar and quality-per-second configuration.

rightmodel/ — 选择合适的模型：使用 Claude Code SKILL 审计 LLM 评估套件，并将其应用于不同的模型和推理参数（扩展思维、算力投入），以找到性价比（质量/美元）和速度（质量/秒）最优的配置。

agent-decomposition/ — Compose Multi-Agent Systems with Skills and MCP: decompose a 400-line-prompt inventory agent into skills + code execution + callable_agents on Claude Managed Agents, with evals to verify each step.

agent-decomposition/ — 使用 Skills 和 MCP 构建多智能体系统：将一个 400 行提示词的库存智能体拆解为 Claude Managed Agents 上的技能（skills）、代码执行（code execution）和可调用智能体（callable_agents），并辅以评估流程来验证每一步。

how-we-claude-code/ — How We Claude Code: a three-phase walkthrough of an AI-assisted product workflow — interview to spec, four divergent design explorations as static HTML, and a Vite + React app whose components emit a machine-readable DOM contract so an agent (or CI) can verify them at runtime.

how-we-claude-code/ — 我们如何进行 Claude Code 开发：AI 辅助产品工作流的三阶段演练——从访谈到规格说明书，生成四种不同的静态 HTML 设计方案，以及一个 Vite + React 应用。该应用的组件会输出机器可读的 DOM 契约，以便智能体（或 CI）在运行时进行验证。

ship-your-first-managed-agent/ — Ship Your First Managed Agent: a Streamlit incident dashboard with an offline SRE Agent chat panel. You bring it online by implementing seven small functions in agent.py, each a single Claude Managed Agents API call — until it can grep a 70k-line log in its sandbox, call your local tools, and name the bad commit.

ship-your-first-managed-agent/ — 发布你的第一个托管智能体：一个带有离线 SRE 智能体聊天面板的 Streamlit 事件仪表盘。你需要通过在 agent.py 中实现七个小函数来使其上线，每个函数对应一个 Claude Managed Agents API 调用——最终它能够在其沙盒中 grep 7 万行日志、调用本地工具并定位出问题的提交（bad commit）。

agent-battle/ — Agent Battle: a 45-minute competition to configure a Claude Managed Agent — system prompt, skills, MCP servers, model — that drives a local game bot over MCP. Most diamonds wins, fewest tokens breaks ties; a fast —eval decision-probe loop lets you test config changes in ~30s before committing to a 5-minute run.

agent-battle/ — 智能体对战：一场 45 分钟的竞赛，要求配置一个 Claude Managed Agent（系统提示词、技能、MCP 服务器、模型）来通过 MCP 驱动一个本地游戏机器人。收集钻石最多者获胜，Token 消耗最少者在平局时胜出；快速的 —eval 决策探测循环让你能在约 30 秒内测试配置更改，然后再进行 5 分钟的正式运行。

agents-that-remember/ — Agents That Remember: start with a Managed Agent that’s visibly amnesiac across sessions, then layer in memory primitives one at a time — a memory store for cross-session persistence, then the Dreaming Service to consolidate past transcripts — going “goldfish to colleague” in 45 minutes.

agents-that-remember/ — 拥有记忆的智能体：从一个在不同会话间表现出“健忘”的托管智能体开始，逐步添加记忆原语——先是用于跨会话持久化的记忆存储，然后是用于整合过往记录的“梦境服务”（Dreaming Service），在 45 分钟内实现从“金鱼记忆”到“可靠同事”的转变。

eval-driven-agent-development/ — Eval-Driven Agent Development: iterate a PPTX-generating Managed Agent through six variants (naive → visual → typography → palette → density → QA-loop), scoring each against a 10-task suite with a two-layer grader (programmatic .pptx XML metrics + LLM-as-judge on rendered slides) so every prompt change is measured, not vibed.

eval-driven-agent-development/ — 评估驱动的智能体开发：通过六个变体（朴素版 → 视觉版 → 排版版 → 配色版 → 密度版 → 质量保证循环版）迭代一个生成 PPTX 的托管智能体，并使用双层评分系统（程序化 .pptx XML 指标 + 基于渲染幻灯片的 LLM 裁判）针对 10 项任务进行评分，确保每一次提示词的修改都是可量化的，而非凭感觉。

production-ready-agent/ — Deal Desk: a chat-first UI over a multi-agent M&A research team on Claude Managed Agents — a coordinator delegates to four parallel research sub-agents, reads prior-deal lessons from a memory store, reaches Linear via MCP, and emits a graded investment thesis while the UI streams every event and gated tool call.

production-ready-agent/ — 交易台（Deal Desk）：一个基于 Claude Managed Agents 的多智能体并购研究团队的聊天优先 UI——协调员将任务委派给四个并行的研究子智能体，从记忆存储中读取过往交易经验，通过 MCP 连接 Linear，并输出分级的投资论文，同时 UI 会实时流式传输每一个事件和受控的工具调用。

research-desk/ — The Research Desk: build an SEC-filings research desk on Claude Managed Agents behind a self-hosted Next.js console — say hello to a bare agent you wire up yourself, then promote that same agent (versioned update) into a head of research that dispatches one analyst session per ticker through a custom tool your server fulfils, with sub-agent specialists, an edgartools Skill, outcome-graded scorecards, a shared memory store, and a weekly memo deployment.

research-desk/ — 研究台：在自托管的 Next.js 控制台后，基于 Claude Managed Agents 构建一个 SEC 文件研究台——从一个你需要亲自配置的原始智能体开始，将其升级（版本更新）为研究主管，通过服务器实现的自定义工具为每个股票代码分派分析师会话，并配备子智能体专家、edgartools 技能、结果评分卡、共享记忆存储以及每周备忘录部署功能。

License Apache License 2.0. See LICENSE.

许可协议：Apache License 2.0。详见 LICENSE 文件。