Signal Garden

Skill 与 AgentSKILLS & AGENTS· MCP-SERVER

ai-agent-book

bojieli/ai-agent-book

AI Agent设计原理与工程实践的开源书籍,含正文、PDF和配套代码。An open-source book with 52k+ stars that teaches AI agent design through hands-on code and experiments.

根据公开文档整理,未运行实测;费用以项目官网为准。Based on public documentation, without runtime testing. Check the project site for pricing.

适用平台Platforms
claude-code · claude-desktop · codex
已识别的要求Documented requirements
API key · Local runtime
许可LICENSE
Apache-2.0

每周一 1 封:本周热度最高的 10 个 GitHub 新项目。免费,随时退订。One email each Monday: the 10 hottest new GitHub projects this week. Free, unsubscribe any time.

订阅即表示你同意我们按隐私政策发送每周一周刊。我们会记录这次订阅的来源页面和活动标签,这条记录不包含 IP 地址。Subscribing means you agree that we email the Monday weekly digest as described in the privacy policy. We record the referring page and campaign tag for the signup, and that record does not include an IP address. 隐私政策Privacy Policy

Featured on Signal Garden
徽章主题Badge theme

能力CAPABILITIES

它能帮你做成什么What it helps you do

  1. 01

    读懂并动手跑全书配套代码和实验

  2. 02

    从失败案例学习循环修复、记忆设计等工程方法

  3. 03

    拿30个现成技巧卡片直接抄进自己的项目

  4. 04

    用一键脚本拉齐各章节外部依赖的固定版本

  1. 01

    Design and evaluate system performance with custom metrics.

  2. 02

    Manage long-term memory and context across interactions.

  3. 03

    Coordinate multiple decision-makers in a shared task.

  4. 04

    Train models with reinforcement learning from feedback.

  5. 05

    Structure prompts and context for better results.

上手GET STARTED

怎么装How to install

GIT-CLONE
git clone https://github.com/anthropics/claude-quickstarts.git chapter6/claude-quickstarts && git -C chapter6/claude-quickstarts checkout --detach 9bcc95e316e5ef6542b4c9d0469f4078829eead5  # 实验 6-8 使用 computer-use-demo/

选择与你的环境对应的方式。命令来自项目文档,未运行实测。Choose the option for your environment. Commands come from project documentation and have not been run-tested.

打开项目文档Open project documentation ↗

使用限制LIMITATIONS

限制与风险Limits and risk

以下要求与风险来自文档静态检查,未执行项目代码;扫描没有命中,不代表使用安全。Risk here is evidence, not an alarm. These findings come from static scanning of the docs; we never execute a project’s code.

限制LIMITS

需检查Needs review

01需要 API KeyREQUIRES API KEY
是Yes
02需要本地环境NEEDS LOCAL RUNTIME
是Yes
03需要 DockerNEEDS DOCKER
未在文档中发现Not found in the docs
04难度DIFFICULTY
需要开发经验Needs development work
05许可LICENSE
Apache-2.0Apache-2.0
06阶段STAGE
持续活跃Active
创建于 09 SEP 2025Created 09 SEP 2025

静态扫描STATIC SCAN

  • 01控制浏览器Controls a browserreadme

    - 浏览器、CUDA、FFmpeg、Ollama、Playwright 浏览器及外部仓库等系统依赖,请继续参考各实验 README。第 8 章部分内置第三方组件需要 Python 3.12+。
  • 02读写本地文件Reads and writes local filesskills/coding-agent-harness/SKILL.md

    1. **工具集**:文件读写(read/write/edit)、目录浏览(glob/ls)、内容搜索(grep)、命令执行(bash)七件套;每个工具命名直观、参数带例子、边界有说明(防呆设计)。
  • 03控制浏览器Controls a browserskills/computer-use/SKILL.md

    - 没有编程 API 可用、Selenium 之类脚本化方案难以维护页面的场景。

静态扫描发现中等风险:项目中的实验会控制浏览器、读写本地文件,并依赖外部仓库(如浏览器自动化、机器人控制等)。这些操作需要你自行评估安全边界,建议在隔离环境中运行实验。The static scan found medium risks for browser control and local file read/write, which are part of the book's experiments. Review the code before running.

项目动态PROJECT ACTIVITY

它现在处在什么位置Where it stands right now

首次发现FIRST SEEN

03 OCT 202603 OCT 2026

增长GROWTH

+13.0%+13.0%

52k → 59k stars · 49 个快照52k → 59k stars · 49 snapshots

状态STATUS

持续活跃Active

质量 A-Quality A-

这本书在 GitHub 上获得了 52592 颗星,说明社区认可度很高。52k+ GitHub stars and 5.9k forks show strong community trust.

每周一周刊Monday weekly digest

每周一 1 封:本周热度最高的 10 个 GitHub 新项目。免费,随时退订。One email each Monday: the 10 hottest new GitHub projects this week. Free, unsubscribe any time.

看第 1 期Read issue 1

订阅即表示你同意我们按隐私政策发送每周一周刊。我们会记录这次订阅的来源页面和活动标签,这条记录不包含 IP 地址。Subscribing means you agree that we email the Monday weekly digest as described in the privacy policy. We record the referring page and campaign tag for the signup, and that record does not include an IP address. 隐私政策Privacy Policy

订阅每周一周刊Subscribe to the Monday weekly digest
PROVENANCE · 这些结论的依据是什么Where these conclusions come from
来源 URLSource URL
https://github.com/bojieli/ai-agent-book
原始数据哈希Raw data hash
47330990bf3d04ec80f0cb15074376ca
解析时间Parsed at
2026-10-04T09:07:17.646Z
最后一次验证Last verified
2026-11-20T22:14:42.723Z
增长测量时间Growth measured
2026-10-04T23:17:28.032Z
发现查询Discovery queries
focus:token:repo:topic-context-engineering
分类来源Category source
rules
焦点Focus
token · design (7)
完整解析证据 JSONFull parse evidence JSON
{
  "agents": [
    {
      "agent": "claude-code",
      "evidence": "readme: …ude 等 LLM 产品;熟悉至少一款 AI 辅助编程工具(Claude Code、Codex、Cursor 等);了解命令行、Git、JSO…"
    },
    {
      "agent": "claude-desktop",
      "evidence": "skills/mcp-skill-hub/SKILL.md: …处处可用。** 一个 MCP 服务器可同时被 Cursor、Claude Desktop、OpenClaw 等任何兼容客户端使用,开发者无需关心上游…"
    },
    {
      "agent": "codex",
      "evidence": "skills/tool-discovery/SKILL.md: …置内联展开 block,官方文档明确后续每轮保持缓存命中);Codex CLI 默认开启 `tool_search`。 - **真正导致重…"
    }
  ],
  "apiKey": "置第三方组件需要 Python 3.12+。 ## 🔑 API 密钥 建议申请下面几个平台的 API Key 方便学习。模型选型可参考 [这篇指南",
  "docker": null,
  "taxonomy": {
    "scores": {
      "design": 3,
      "ai-tools": 7,
      "dev-tools": 0,
      "indie-web": 0,
      "skill-agent": 3,
      "productivity": 3
    },
    "primary": "skill-agent",
    "secondary": "mcp-server"
  },
  "localRuntime": "环境变量。只有在实验 README 或 CLI 明确列出 `ollama` 时,才可启动本地 Ollama 并添加 `--provider ollama",
  "skillMdTotal": 30,
  "skillMdErrors": [
    {
      "path": "chapter8/curly-quote-sft/SKILL.md",
      "error": "no frontmatter"
    }
  ],
  "treeTruncated": false,
  "categoryScores": {
    "data": 0,
    "agent": 3,
    "media": 0,
    "design": 3,
    "browser": 0,
    "devtool": 0,
    "security": 0,
    "marketing": 0,
    "productivity": 3
  },
  "classification": [
    {
      "signal": "29 SKILL.md with name + description",
      "weight": 0.95
    },
    {
      "signal": "python dependency on mcp / fastmcp + MCP framing",
      "weight": 0.8
    },
    {
      "signal": "skills/ directory with 23 markdown files",
      "weight": 0.75
    },
    {
      "signal": "README describes a Claude / Agent Skill",
      "weight": 0.7
    },
    {
      "signal": "description mentions AI",
      "weight": 0.2
    }
  ],
  "scannedSources": [
    "readme",
    "chapter10/multi-role-transfer/skills/coding/SKILL.md",
    "chapter10/multi-role-transfer/skills/data_analysis/SKILL.md",
    "chapter10/multi-role-transfer/skills/research/SKILL.md",
    "chapter10/multi-role-transfer/skills/triage/SKILL.md",
    "chapter10/multi-role-transfer/skills/writing/SKILL.md",
    "chapter2/agent-skills-ppt/skills/pptx/SKILL.md",
    "chapter8/curly-quote-sft/SKILL.md",
    "chapter9/ai-style-skill/skill/SKILL.md",
    "skills/agent-evaluation/SKILL.md",
    "skills/agent-evolution/SKILL.md",
    "skills/agent-state-bar/SKILL.md",
    "skills/async-event-agent/SKILL.md",
    "skills/bad-case-to-dpo/SKILL.md",
    "skills/coding-agent-harness/SKILL.md",
    "skills/computer-use/SKILL.md",
    "skills/context-compression/SKILL.md",
    "skills/context-engineering/SKILL.md",
    "skills/error-recovery/SKILL.md",
    "skills/eval-dataset-design/SKILL.md",
    "skills/knowledge-org/SKILL.md",
    "skills/kv-cache-design/SKILL.md",
    "skills/loop-engineering/SKILL.md",
    "skills/mcp-skill-hub/SKILL.md",
    "skills/memory-system/SKILL.md",
    "skills/multi-agent-design/SKILL.md",
    "skills/post-training-strategy/SKILL.md",
    "skills/rag-pipeline/SKILL.md",
    "skills/reward-design/SKILL.md",
    "skills/tool-design/SKILL.md",
    "skills/tool-discovery/SKILL.md"
  ],
  "skillMdFetched": 30
}