# Wayback

> Wayback is a freemium MCP server for web search and scraping. MCP server for querying Internet Archive Wayback Machine snapshots and capture history. It can run locally on your machine or as a hosted remote server.

- **Kind:** MCP server
- **Pricing:** Freemium
- **Starting price:** $49 a month
- **Platforms:** API, CLI, self-hosted
- **Website:** https://pipeworx.io/packs/wayback
- **Repository:** https://github.com/pipeworx-io/mcp-wayback
- **MCP Registry name:** io.github.pipeworx-io/wayback
- **Categories:** [web search and scraping](https://ai-powered.com/categories/web-search-and-scraping)
- **Listing updated:** 2026-11-18

## About

The README describes this as an MCP wrapper around the Internet Archive's Wayback Machine, exposing three lookups: the nearest archived copy of a page (optionally at a given date), a dated list of captures from the CDX index with optional from/to ranges, and a summary of how many times a URL was captured along with its first and last capture times.

The server is published on npm as a stdio bridge that forwards calls to a hosted Pipeworx gateway endpoint, and the gateway also offers a streamable-HTTP connection. An optional Pipeworx API key raises the anonymous daily call cap; without it calls are limited per day.

The package is aimed at developers and agent builders who need historical versions of web pages or want to check when a site was archived, without writing their own CDX queries. The gateway also lists shared meta-tools alongside the three pack tools.

## Features

- **Closest snapshot lookup:** Returns the nearest archived version of a URL with its Wayback URL, timestamp and HTTP status, optionally for a chosen date.
- **Capture history listing:** Lists captures over time from the CDX index with timestamp, status code, MIME type and content digest, filterable by date range.
- **Capture count summary:** Reports the total number of captures for a URL and its earliest and latest capture timestamps.
- **Stdio and HTTP transports:** Can be run locally via npx over stdio or connected to the hosted endpoint over streamable HTTP.
- **Optional API key:** An optional Pipeworx key raises the daily call allowance; calls are metered when a key is used.

## Good for

- Retrieve an archived copy of a page that has since changed or disappeared
- Check when a domain was first and last captured by the Internet Archive
- Audit how a URL's content or HTTP status changed across dated captures
- Let an AI assistant answer questions about a site's historical versions

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http wayback https://gateway.pipeworx.io/wayback/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://gateway.pipeworx.io/wayback/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "wayback": {
      "url": "https://gateway.pipeworx.io/wayback/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "wayback": {
      "type": "http",
      "url": "https://gateway.pipeworx.io/wayback/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "wayback": {
      "serverUrl": "https://gateway.pipeworx.io/wayback/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "wayback": {
      "type": "http",
      "url": "https://gateway.pipeworx.io/wayback/mcp"
    }
  }
}
```

## Tools

- `get_snapshot`: Look up the archived version of a web page, optionally nearest a timestamp.
- `list_snapshots`: List captured snapshots of a URL over time from the CDX index, with optional from/to dates.
- `get_capture_count`: Count captures of a URL and return its first and last capture timestamps.

## FAQ

### Which tools does the Wayback server provide?

The README lists get_snapshot, list_snapshots and get_capture_count. The hosted endpoint also lists shared Pipeworx meta-tools such as ask_pipeworx.

### Do I need an API key?

No. Calls work anonymously with a daily cap. A free Pipeworx account raises the cap to 200 calls a day.

### Can the npm package list only the Wayback tools?

Yes. By default it lists only this pack's tools; setting PIPEWORX_META_TOOLS=1 also lists the gateway-wide tools.

### Is it affiliated with the Internet Archive?

The README states it is an independent, unofficial integration and not endorsed by the upstream provider.

### Is Wayback free?

Wayback has a free plan, with paid plans for more usage or features. Paid plans start at $49 a month.

### How do I add Wayback to Claude Code?

Run this command in your terminal: claude mcp add --transport http wayback https://gateway.pipeworx.io/wayback/mcp

### What are the best alternatives to Wayback?

The closest alternatives to Wayback are Ipify, Wikimedia Commons MCP and Unsplash.

## Where this comes from

- Description: written from the project's README, the project's website and its MCP Registry entry.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [Ipify](https://ai-powered.com/mcp/ipify): Ipify is a freemium MCP server for web search and scraping.
- [Wikimedia Commons MCP](https://ai-powered.com/mcp/wikimedia-commons-mcp): Wikimedia Commons MCP is an open-source MCP server for web search and scraping.
- [Unsplash](https://ai-powered.com/mcp/unsplash): Unsplash is a freemium MCP server for web search and scraping.

---
Source: https://ai-powered.com/mcp/wayback
