mcp-server-wayback 0.2.1

MCP server giving Claude and other LLM clients structured access to the Internet Archive's Wayback Machine.

wayback-mcp

A Model Context Protocol server giving Claude structured access to the Internet Archive's Wayback Machine.

---

Overview

wayback-mcp is an async Python MCP server that exposes the Internet Archive's six core APIs — Availability, CDX, Advanced Search, Metadata, and Wayback content — as first-class tools, prompts, and resources for Claude. It handles rate limiting, retry/back-off, and response shape normalisation so the model only sees structured Pydantic data.

Features

• Six MCP tools covering availability checks, snapshot lookups, full-text item search, domain crawls, page-text extraction, and item metadata
• Four guided prompts — research_topic, track_site_changes, audit_link_rot, setup_authentication
• One MCP resource — wayback://item/{identifier} exposes IA item metadata as JSON
• Async token-bucket rate limiter with per-endpoint buckets and Retry-After honoring
• In-memory response cache with per-endpoint TTLs to keep token usage and IA load low
• Internet Archive S3 authentication (optional) for higher rate-limit ceilings
• Structured error model — expected failures return ToolError; unexpected ones raise
• Tested against live IA APIs via an opt-in --integration pytest flag

Quick start

One command, then restart Claude Desktop:

uvx mcp-server-wayback --install

That's it. The installer writes the wayback entry into your claude_desktop_config.json (merging with anything already there) and prints a restart hint. To remove it later: uvx mcp-server-wayback --uninstall.

Need uvx? brew install uv on macOS, or pipx install uv. Python 3.11+ required.

Other clients

• Claude Code: claude mcp add wayback -- uvx mcp-server-wayback
• Anything else (Cursor, Windsurf, Codex CLI, Zed, etc.): most clients accept an MCP config block — use this one:

{
  "mcpServers": {
    "wayback": {
      "command": "uvx",
      "args": ["mcp-server-wayback"]
    }
  }
}

Optional: Internet Archive authentication

Set both keys in the server's environment to authenticate every IA request and raise your rate-limit ceiling. Run the setup_authentication prompt from Claude to walk through it interactively.

"env": {
  "WAYBACK_MCP_IA_ACCESS_KEY": "<your access key>",
  "WAYBACK_MCP_IA_SECRET_KEY": "<your secret key>"
}

Get keys at https://archive.org/account/s3.php.

Tools

Tool	Purpose
check_availability	Is this URL archived? Returns the closest snapshot
lookup_snapshots	List CDX snapshots for a URL with date / status filters
search_archive	Lucene search across IA collections with mediatype + year range
search_domain	Discover archived URLs under a domain or path prefix
get_snapshot_content	Fetch an archived page and extract its readable text
get_item_metadata	Rich structured metadata for any IA item identifier

Prompts

Prompt	What it does
research_topic	Multi-mediatype IA search → synthesised topic overview
track_site_changes	Sample snapshots over time → narrate how a page evolved
audit_link_rot	Bulk-check URLs and surface archived alternatives
setup_authentication	Walks the user through configuring IA S3 keys

Development

Requires Python 3.11+ and uv.

git clone https://github.com/lakshyamehta03/wayback-machine-mcp.git
cd wayback-machine-mcp
uv sync
uv run mcp-server-wayback      # run the server
uv run pytest                  # unit tests (httpx mocked via respx)
uv run pytest --integration    # also hit live Internet Archive APIs

CI runs the unit suite on every push and pull request via GitHub Actions.

License

MIT. The Wayback Machine logo is © Internet Archive and used here under fair use to identify the upstream service this project integrates with.

Acknowledgments

• The Internet Archive for the Wayback Machine and the open APIs that make this server possible
• Anthropic for the Model Context Protocol specification and SDK