Wayback Machine (archive.org) MCP server for AI agents

Let Claude, ChatGPT or your own agent read Wayback Machine (archive.org): 8 typed functions through one MCP server. No Wayback Machine (archive.org) API key needed.

Functions
8
Sign-in
Not needed
Works with
MCP, HTTP, Node, Python

Connect

Just Wayback Machine (archive.org)

An MCP server for Wayback Machine (archive.org) alone, with one tool per function. Paste it into Claude, ChatGPT, Cursor or any MCP client as a remote server.

https://api.bowmark.ai/mcp/site/archive_org
Every site

One MCP server for Wayback Machine (archive.org) and every other site Bowmark covers. Your agent looks up what it needs and calls it.

https://api.bowmark.ai/mcp
Set it up in your client

Both need a Bowmark API key. The quickstart gets you one and a first result in a few minutes.

Functions

Open a function to see its arguments, what it returns and how to call it.

Read 8
  • checkAvailabilityThe Wayback Machine's own public availability lookup.

    Arguments

    NameTypeRequired
    sitestringYes
    timestampstringNo

    Returns archive_orgAvailability

    interface archive_orgAvailability {
      query: string;
      available: boolean;
      archivedUrl: string | null;
      archivedTimestamp: string | null;  // YYYYMMDDhhmmss
      archivedStatus: string | null;     // the original page's own HTTP status when captured
    }

    HTTP endpoint

    POST https://api.bowmark.ai/v1/sites/archive_org/checkAvailability

    Call it

    import { bowmark } from "@bowmark/web";
    
    const result = await bowmark.providers.archive_org.checkAvailability("cua.ai");
  • checkLendingAvailabilityChecks whether a book item is available to borrow from the Internet Archive right now.

    Arguments

    NameTypeRequired
    identifierstringYes

    Returns archive_orgLendingAvailability

    interface archive_orgLendingAvailability {
      identifier: string;
      isLendable: boolean;             // false for a public-domain item nothing controls lending on
      isReadable: boolean;             // true when the item is readable outright (public domain or immediately available)
      availableToBorrow: boolean;      // false when not lendable at all, OR when every copy is checked out, see waitlistSize
      availableToBrowse: boolean;
      availableBorrowableCopies: number;
      maxBorrowableCopies: number;
      waitlistSize: number;
    }

    HTTP endpoint

    POST https://api.bowmark.ai/v1/sites/archive_org/checkLendingAvailability

    Call it

    import { bowmark } from "@bowmark/web";
    
    const result = await bowmark.providers.archive_org.checkLendingAvailability("prideprejudice00aust");
  • downloadFileBuilds a download URL for one file within an item.

    Arguments

    NameTypeRequired
    identifierstringYes
    filenamestringYes

    Returns archive_orgDownloadUrl

    interface archive_orgDownloadUrl {
      filename: string;                // the file name within the item
      downloadUrl: string;             // the ready-to-fetch URL for this file
    }

    HTTP endpoint

    POST https://api.bowmark.ai/v1/sites/archive_org/downloadFile

    Call it

    import { bowmark } from "@bowmark/web";
    
    const result = await bowmark.providers.archive_org.downloadFile("prideprejudice00aust", "prideprejudice00aust.pdf");
  • getItemFetches one item's full metadata.

    Arguments

    NameTypeRequired
    identifierstringYes

    Returns archive_orgItem

    interface archive_orgFile {
      name: string;                 // pass to downloadFile, or use downloadUrl directly
      format: string | null;        // e.g. "EPUB", "Text PDF", "Animated GIF"
      size: number | null;          // bytes
      downloadUrl: string;
    }
    
    interface archive_orgItem {
      identifier: string;
      title: string | null;
      creator: string[] | null;
      description: string | null;
      date: string | null;
      mediatype: string | null;
      collection: string[] | null;
      files: archive_orgFile[];
    }

    HTTP endpoint

    POST https://api.bowmark.ai/v1/sites/archive_org/getItem

    Call it

    import { bowmark } from "@bowmark/web";
    
    const result = await bowmark.providers.archive_org.getItem("prideprejudice00aust");
  • getSnapshotReads one archived page as it was captured.

    Arguments

    NameTypeRequired
    urlstringYes
    timestampstringYes

    Returns archive_orgSnapshotPage

    interface archive_orgSnapshotPage {
      originalUrl: string;
      timestamp: string;
      archivedUrl: string;
      html: string;                 // the ORIGINAL bytes as captured, without the archive's toolbar or rewritten links
      warnings: string[];
    }

    HTTP endpoint

    POST https://api.bowmark.ai/v1/sites/archive_org/getSnapshot

    Call it

    import { bowmark } from "@bowmark/web";
    
    const result = await bowmark.providers.archive_org.getSnapshot("<url>", "<timestamp>");
  • listSnapshotsLists the Wayback Machine's captures, newest first.

    Arguments

    NameTypeRequired
    sitestringYes
    opts.scope"exact" | "prefix"
    Default: "prefix" for a bare domain (every page under it), "exact" for a url with a path.
    No
    opts.limitnumber
    1-500, default 50, newest first.
    No
    opts.pathContainsstring[]
    Keep only urls whose path contains one of these, e.g. ["about", "team", "contact"].
    No

    Returns archive_orgSnapshotList

    interface archive_orgSnapshot {
      timestamp: string;            // YYYYMMDDhhmmss
      originalUrl: string;
      archivedUrl: string;          // the browsable capture, web.archive.org/web/<ts>/<url>
      statusCode: string;
      mimeType: string;
    }
    
    interface archive_orgSnapshotList {
      query: string;
      scope: "exact" | "prefix";
      snapshots: archive_orgSnapshot[];   // prefix: newest capture of each distinct url; exact: each distinct VERSION of the page
      warnings: string[];
    }

    HTTP endpoint

    POST https://api.bowmark.ai/v1/sites/archive_org/listSnapshots

    Call it

    import { bowmark } from "@bowmark/web";
    
    const result = await bowmark.providers.archive_org.listSnapshots("<site>");
  • searchItemsSearch the Internet Archive's library of books, films, audio, software and more using the Solr search API.

    Arguments

    NameTypeRequired
    querystringYes
    opts.mediatypestring
    Narrow to one Solr mediatype, e.g. "texts", "audio", "movies", "software", "image".
    No
    opts.sort"relevance" | "downloads" | "date"
    Default: "relevance".
    No
    opts.limitnumber
    1-200, default 20.
    No

    Returns archive_orgSearchResults

    interface archive_orgSearchItem {
      identifier: string;           // pass to getItem / checkLendingAvailability / downloadFile
      title: string | null;
      creator: string[] | null;
      date: string | null;
      mediatype: string | null;
      downloads: number | null;
    }
    
    interface archive_orgSearchResults {
      query: string;
      totalFound: number;           // Solr's numFound, the full match count, not just items.length
      items: archive_orgSearchItem[];
      warnings: string[];
    }

    HTTP endpoint

    POST https://api.bowmark.ai/v1/sites/archive_org/searchItems

    Call it

    import { bowmark } from "@bowmark/web";
    
    const result = await bowmark.providers.archive_org.searchItems("pride and prejudice");
  • searchScholarSearches IA Scholar (scholar.archive.org), the Internet Archive's full-text index of open-access research papers.

    Arguments

    NameTypeRequired
    querystringYes
    opts.pagenumber
    1-based, 20 results a page; default 1.
    No

    Returns archive_orgScholarResults

    interface archive_orgScholarResult {
      title: string;
      authors: string[];
      year: number | null;
      venue: string | null;            // journal, book or conference
      snippet: string;                 // full-text highlights around the query terms
      doi: string | null;
      externalIds: string[];           // e.g. "doi:10.1038/nbt.2863", "pmid:24714470"
      fulltextUrl: string | null;      // the archived PDF/full text on scholar.archive.org
      fulltextLabel: string | null;    // e.g. "Archived PDF (342 kB, captured 2017)"
      originalUrl: string | null;      // the publisher's own url (often doi.org)
      permalink: string | null;
      releaseUrl: string | null;       // the catalog record behind the result
    }
    
    interface archive_orgScholarResults {
      query: string;
      page: number;
      totalHits: string | null;        // as the site prints it, e.g. "10,000+"
      hasNextPage: boolean;
      results: archive_orgScholarResult[];
      warnings: string[];
    }

    HTTP endpoint

    POST https://api.bowmark.ai/v1/sites/archive_org/searchScholar

    Call it

    import { bowmark } from "@bowmark/web";
    
    const result = await bowmark.providers.archive_org.searchScholar("crispr gene editing");

Use it from code