Internet Archive Mcp Server
Search the Wayback Machine and Internet Archive library from your agent when researching niches, dead competitors, and primary sources.
Overview
io.github.cyanheads/internet-archive-mcp-server is an Idea-phase MCP server that searches the Wayback Machine and Internet Archive library and returns snapshots, metadata, and text.
What is this MCP server?
- Search Wayback Machine and Internet Archive library (40M+ items)
- Fetch snapshots, item metadata, and text content
- Stdio and local streamable-http transport options (v0.1.2)
- Bun/npm package @cyanheads/internet-archive-mcp-server
- MCP_LOG_LEVEL configuration
- Internet Archive library described as 40M+ items
- Package version 0.1.2
- Stdio and streamable-http transports
What problem does it solve?
You need Wayback and Internet Archive evidence inside agent sessions without manually juggling archive.org tabs and snapshot URLs.
Who is it for?
Builders researching market history, deprecated competitor sites, or archival content for blogs and documentation products.
Skip if: Builders who only need live-page scraping of current sites—use a live HTTP or browser MCP instead of archival search.
What do I get? / Deliverables
Your agent can search 40M+ IA items, pull Wayback snapshots, and return metadata and text for research and positioning work.
- Wayback snapshot lookups and IA library search results
- Item metadata and text payloads for agent summarization
Recommended MCP Servers
Journey fit
Historical web snapshots and IA metadata are core idea-phase research inputs before you commit to positioning or features. Research subphase fits unified search across Wayback snapshots, item metadata, and extracted text from a 40M+ item corpus.
How it compares
Internet Archive and Wayback MCP connector, not a general-purpose Google search or SERP skill.
Common Questions / FAQ
Who is internet-archive-mcp-server for?
Indie builders and researchers using MCP agents who rely on Wayback Machine and Internet Archive collections during early discovery.
When should I use internet-archive-mcp-server?
Use it when you need historical site snapshots, IA item records, or extracted text while researching a niche or validating positioning claims.
How do I add internet-archive-mcp-server to my agent?
Install @cyanheads/internet-archive-mcp-server from npm, run start:stdio for stdio MCP, or configure the streamable-http URL on localhost per server.json.