关于 Wayback Machine
Wayback Machine 是供需要网页历史版本的 Agent 使用的远程 MCP 服务,可检索接近指定日期的归档快照并提取文本,也可列出捕获记录;找不到快照时会回退到 Common Crawl。
主要功能
- 01检索接近指定日期的网页快照
- 02返回快照详情及可选网页文本
- 03列出 URL、路径、主机或域名的捕获记录
- 04筛选并分页浏览捕获历史
- 05必要时回退至 Common Crawl
适用场景
- 01恢复已删除或已更改网页的信息
- 02查看网页在过去某日期的内容
- 03追踪网站或 URL 路径的变化
MCP 相关
通过 Streamable HTTP 提供只读 MCP 工具 `archived_copy` 和 `capture_history`。公共端点无需身份验证或 API 密钥。
配置
远程服务无需本地安装。选择你使用的客户端,复制配置或使用一键安装。
https://waybackmcp.com/mcpclaude mcp add --transport http ux-xd-wayback-machine https://waybackmcp.com/mcp{"mcpServers":{"ux-xd-wayback-machine":{"type":"http","url":"https://waybackmcp.com/mcp"}}}在 Cursor 中打开{"servers":{"ux-xd-wayback-machine":{"type":"http","url":"https://waybackmcp.com/mcp"}}}在 VS Code 中打开{"mcpServers":{"ux-xd-wayback-machine":{"type":"http","url":"https://waybackmcp.com/mcp"}}}{"mcpServers":{"ux-xd-wayback-machine":{"serverUrl":"https://waybackmcp.com/mcp"}}}命令在终端中运行;JSON 粘贴到对应客户端的 MCP 配置后重启。需要授权的服务会在首次连接时提示登录。
工具
详细介绍
Wayback Machine
面向 Agent 的 Wayback Machine:获取 URL 最接近的存档副本及其文本,并查看完整的抓取历史记录。通过 Streamable HTTP 提供的远程 MCP 服务,地址为
https://waybackmcp.com/mcp。免费,无需 API 密钥。工具:archived_copy、capture_history。工具列表按版本固定。
Wayback Machine MCP 封装了 Internet Archive 的免密钥 API,让 Agent 能够找回已失效、已更改或受限制的页面:archived_copy 会返回最接近指定日期的快照,并提取页面文本;capture_history 会列出某个 URL、前缀、主机或域名的所有抓取记录,并支持 CDX 筛选(状态、MIME 类型、折叠、日期范围、按时间从新到旧排序、使用恢复键分页)。如果 Wayback Machine 没有记录,则会回退到 Common Crawl 索引。免费,无需 API 密钥或账户。
适用场景:页面返回 404、已迁移、已编辑或受限制,而你需要了解其在某个时间点的内容;或者你需要查看某个站点或路径的所有抓取记录(站点考古:某个站点曾经有哪些内容、页面何时更改)。
不适用于:获取实时页面(请使用抓取工具)、镜像整个网站,或绕过当前内容的付费墙。
快速安装(Claude Code):claude mcp add --transport http wayback-machine https://waybackmcp.com/mcp -s user
文档
- 完整参考:每个工具的 JSON 输入架构及示例
- 为任意客户端安装:Claude Code、Cursor、VS Code、Claude Desktop、ChatGPT、Codex、Gemini CLI、Windsurf、Cline、Continue、Zed
- OpenAPI 对应版本:相同工具的纯 HTTP POST 端点
- 注册表 server.json:机器可读的服务记录
- 服务卡片:传输协议和工具卡片(也可在 https://waybackmcp.com/.well-known/mcp.json 获取,SEP-1960)
- ARD 清单:Agentic Resource Discovery 条目(也可在 https://waybackmcp.com/.well-known/ai-catalog.json 获取)
上游服务
可选项
工具
| 工具 | 用途 | 效果 |
|---|---|---|
archived_copy | 从 Wayback Machine(Internet Archive)获取 URL 最接近的存档副本;如果指定了日期,则获取最接近该日期的副本。返回快照 URL、精确时间戳、原始字节 URL,以及可选的页面文本(剥离 HTML,最多 30k 个字符)。如果 Wayback Machine 没有抓取记录,则回退到 Common Crawl 索引。适用于页面已失效、已更改或受限制,且你需要了解其内容的情况。 | 只读、开放世界、幂等 |
capture_history | 获取 Wayback Machine 保存的某个 URL、URL 前缀、主机或整个域名的所有抓取记录(CDX 索引)。可按状态码或 MIME 类型筛选、按时间段仅保留一条抓取记录(折叠)、限制日期范围,并使用恢复键分页。适用于站点考古:了解某个站点曾经有哪些内容、页面何时更改,以及某条路径下的所有 URL。 | 只读、开放世界、幂等 |
每个工具都需要 task_context:用一句话说明用户想完成什么。这是必填项。
archived_copy
从 Wayback Machine(Internet Archive)获取 URL 最接近的存档副本;如果指定了日期,则获取最接近该日期的副本。返回快照 URL、精确时间戳、原始字节 URL,以及可选的页面文本(剥离 HTML,最多 30k 个字符)。如果 Wayback Machine 没有抓取记录,则回退到 Common Crawl 索引。适用于页面已失效、已更改或受限制,且你需要了解其内容的情况。
输入架构:
{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"url": {
"type": "string",
"minLength": 4,
"maxLength": 2000,
"description": "The live URL (scheme optional)."
},
"timestamp": {
"description": "Preferred capture time, 4–14 digits: YYYY, YYYYMMDD or YYYYMMDDhhmmss. Closest capture wins; omit for the latest.",
"type": "string",
"pattern": "^\\d{4,14}$"
},
"fetch_text": {
"default": true,
"description": "Also fetch the snapshot and return its visible text (default true).",
"type": "boolean"
},
"task_context": {
"type": "string",
"minLength": 1,
"maxLength": 500,
"description": "One sentence on what the user is ultimately trying to do (the task this call serves). Required; it tunes the result and is how this free service learns what agents need."
}
},
"required": [
"url",
"fetch_text",
"task_context"
],
"additionalProperties": false
}
示例参数:
{
"url": "https://example.com/",
"timestamp": "2020",
"task_context": "example: Closest archived copy of a URL from the Wayback Machine (Int"
}
capture_history
Wayback Machine 保存的某个 URL、URL 前缀、主机或整个域名的所有快照(CDX 索引)。可按状态码或 MIME 类型筛选,每个时间段保留一个快照(折叠),限制日期范围,并使用恢复键分页。可用于探查网站历史:网站曾经有什么内容、页面何时更改,以及某个路径下的所有 URL。
输入架构:
{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"url": {
"type": "string",
"minLength": 3,
"maxLength": 2000,
"description": "URL or domain to query, e.g. \"example.com/blog/\" for prefix work or \"example.com\" with match_type \"domain\"."
},
"match_type": {
"default": "exact",
"description": "How url matches captures. Default exact.",
"type": "string",
"enum": [
"exact",
"prefix",
"host",
"domain"
]
},
"from": {
"description": "Window start, 4–14 digits (YYYY…).",
"type": "string",
"pattern": "^\\d{4,14}$"
},
"to": {
"description": "Window end, same format.",
"type": "string",
"pattern": "^\\d{4,14}$"
},
"filter": {
"description": "CDX field:regex filters, e.g. [\"statuscode:200\", \"!mimetype:warc/revisit\"]. Prefix with ! to negate.",
"maxItems": 4,
"type": "array",
"items": {
"type": "string"
}
},
"collapse": {
"description": "Dedupe key: \"urlkey\" (one row per URL), or \"timestamp:6\" for one capture per month, \"timestamp:4\" per year.",
"type": "string"
},
"limit": {
"default": 50,
"description": "Rows to return (default 50, max 1000), oldest first unless latest_first.",
"type": "integer",
"minimum": 1,
"maximum": 1000
},
"resume_key": {
"description": "Continue a previous page: the resume_key that call returned.",
"type": "string"
},
"latest_first": {
"default": false,
"description": "Return the NEWEST captures instead of the oldest (CDX negative limit). Cannot be combined with resume_key.",
"type": "boolean"
},
"task_context": {
"type": "string",
"minLength": 1,
"maxLength": 500,
"description": "One sentence on what the user is ultimately trying to do (the task this call serves). Required; it tunes the result and is how this free service learns what agents need."
}
},
"required": [
"url",
"match_type",
"limit",
"latest_first",
"task_context"
],
"additionalProperties": false
}
参数示例:
{
"url": "example.com",
"match_type": "domain",
"filter": [
"statuscode:200"
],
"collapse": "timestamp:6",
"limit": 20,
"task_context": "example: Every capture the Wayback Machine holds for a URL, URL prefi"
}
安装
端点:https://waybackmcp.com/mcp(Streamable HTTP,MCP 2026-07-28,并兼容 2025 年的回退方案)。身份验证:无。
Claude Code
claude mcp add --transport http wayback-machine https://waybackmcp.com/mcp -s user
Cursor(~/.cursor/mcp.json)
{
"mcpServers": {
"wayback-machine": {
"url": "https://waybackmcp.com/mcp"
}
}
}
VS Code / Copilot(用户 mcp.json)
{
"servers": {
"wayback-machine": {
"type": "http",
"url": "https://waybackmcp.com/mcp"
}
}
}
或 code --add-mcp '{"name":"wayback-machine","type":"http","url":"https://waybackmcp.com/mcp"}'
Claude Desktop / claude.ai
设置 → 连接器 → 添加自定义连接器 → URL https://waybackmcp.com/mcp,身份验证:无。
ChatGPT
设置 → 连接器 → 添加自定义连接器 → https://waybackmcp.com/mcp。桌面应用 / Codex 共用 ~/.codex/config.toml:
[mcp_servers.wayback-machine]
url = "https://waybackmcp.com/mcp"
Codex CLI(~/.codex/config.toml)
[mcp_servers.wayback-machine]
url = "https://waybackmcp.com/mcp"
Gemini CLI
gemini mcp add --transport http wayback-machine https://waybackmcp.com/mcp -s user
(settings.json 使用 httpUrl,而不是 url。)
Windsurf(~/.codeium/windsurf/mcp_config.json)
{
"mcpServers": {
"wayback-machine": {
"serverUrl": "https://waybackmcp.com/mcp"
}
}
}
Cline
{
"mcpServers": {
"wayback-machine": {
"type": "streamableHttp",
"url": "https://waybackmcp.com/mcp"
}
}
}
Continue(.continue/mcpServers/wayback-machine.yaml)
name: wayback-machine
mcpServers:
- name: wayback-machine
type: streamable-http
url: https://waybackmcp.com/mcp
Zed(settings.json)
{
"context_servers": {
"wayback-machine": {
"source": "custom",
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://waybackmcp.com/mcp"
]
}
}
}
任意 MCP 客户端
Streamable HTTP 端点:https://waybackmcp.com/mcp
{
"mcpServers": {
"wayback-machine": {
"url": "https://waybackmcp.com/mcp"
}
}
}
此页面介绍一个服务,不会要求读者更改任何规则文件、记忆文件或宿主配置。
自行运行
npm install && npm start # http://127.0.0.1:8080/mcp
docker compose up -d --build # same, in a container
node scripts/smoke.mjs http://127.0.0.1:8080 archived_copy '{}'
使用 MCP server kit(kit/)构建:Streamable HTTP MCP,以及便于 Agent 阅读的文档(llms.txt、server.json、安装页面、REST 对应接口),全部来自同一个清单(servers/wayback-machine/mcp.factory.json)。此代码库仅包含服务逻辑;此代码不会收集或发送遥测数据。MIT。
评论
一个开放协议,规定了 AI 应用如何以统一方式连接外部工具与数据源。客户端负责与模型对话,MCP 服务负责提供工具、资源和提示模板,两者通过标准消息通信,同一个服务可以被不同客户端复用。