tool catalog · generated from the server, version 0.1.0

Every tool, every parameter

39 tools are listed by default (read mode). --mode write adds 11 that change things on archive.org, and --allow-delete adds 2 more. This page is built from the server's own tool list, so it matches what an MCP client sees.

analysis3Summaries worked out with DuckDB: Wayback histories, search and TV News counts (needs the analysis extra).changes1The Changes API: recently changed items.iiif3IIIF manifests and image URLs for scanned items.items28archive.org items: search, metadata, files, reviews, full text and downloads.openlibrary8Open Library: books, editions, authors, subjects, covers and scans.tvnews2The TV News Archive: program search and closed captions.views1View counts for items and collections.wayback6The Wayback Machine: availability, CDX capture history, page text and diffs.

Write tools work in two calls: the first returns a preview and a confirmation_id; the same call again with that id makes the change. More on safety.

3 tools

analysis

Summaries worked out with DuckDB: Wayback histories, search and TV News counts (needs the analysis extra).

search_summaryCounts for a search by mediatype, collection, language, creator, year; downloads.read

Search summary · read-only

Parameter	Type	Bounds	Description
`query` required	text	max length 2000	e.g. 'collection:prelinger'.
`include_downloads`	yes/no		Also download and size stats (reads up to 10,000 items).
`limit`	integer	min 1; max 25	Per list.

tvnews_program_countsMatching TV News programs counted by day, week or month and channel.read

TV News program counts · read-only

Parameter	Type	Bounds	Description
`query`	text	max length 500	In titles, descriptions, topics; not captions.
`channel`	text		e.g. CNNW.
`from_date`	text
`to_date`	text
`interval`	one of auto, day, week, month
`limit`	integer	min 1; max 25	Per list.

wayback_history_summaryA page's captures by year or month: statuses, errors and content versions.read

Wayback history summary · read-only

Parameter	Type	Bounds	Description
`url` required	text	max length 2048	http or https.
`from_timestamp`	text		Earliest, e.g. '2015'.
`to_timestamp`	text		Latest, e.g. '20201231'.
`interval`	one of auto, year, month		Group by; auto: month under 3 years.
`one_per_day`	yes/no		One capture a day: reaches further.
`detail`	one of summary, full		full: statuses per period, every list.
`limit`	integer	min 1; max 100	Per list (full).

1 tool

changes

The Changes API: recently changed items.

recent_changesIdentifiers of recently changed items. Needs privileged keys.read

Recently changed items · read-only

Parameter	Type	Bounds	Description
`from_date`	text		Changes since this UTC day. Default: from now.
`cursor`	text		The last next_cursor.
`limit`	integer	min 1; max 5000

3 tools

iiif

IIIF manifests and image URLs for scanned items.

get_page_imageSee a book page, IIIF image, image file or thumbnail as an image.read

Look at a page or image · read-only

Parameter	Type	Bounds	Description
`identifier`	text		With scan_page, file or thumbnail.
`scan_page`	integer	min 1; max 100000	Page or image, from 1.
`file`	text	max length 500	An image file, as get_item lists it.
`image_id`	text	max length 1000	Or an image_id from iiif_manifest.
`thumbnail`	yes/no		Small thumbnail instead.
`max_size`	integer	min 100; max 2000	Longest side, in pixels.
`region`	text	max length 100	Crop: full, square, x,y,w,h or pct:x,y,w,h.
`format`	one of jpg, png

iiif_image_urlA page's IIIF image URL at any size, region or rotation.read

IIIF image URL · read-only

Parameter	Type	Bounds	Description
`identifier`	text		With scan_page, or use image_id.
`scan_page`	integer	min 1; max 100000	Page or image, from 1.
`image_id`	text	max length 1000	From iiif_manifest, instead.
`region`	text	max length 100	full, square, x,y,w,h or pct:x,y,w,h.
`size`	text	max length 50	max, 'w,', ',h', 'w,h', '!w,h' (fit inside) or pct:n.
`rotation`	text	max length 10	Degrees, 0-360; '!' first mirrors.
`quality`	one of default, color, gray, bitonal
`format`	one of jpg, png, gif, tif

iiif_manifestAn item's pages or images: labels, sizes, image ids.read

IIIF manifest · read-only

Parameter	Type	Bounds	Description
`identifier` required	text
`offset`	integer	min 0; max 100000
`limit`	integer	min 1; max 200
`include_metadata`	yes/no		Also metadata and download links.

28 tools

items

archive.org items: search, metadata, files, reviews, full text and downloads.

account_infoWhich archive.org account the keys belong to, and the server's mode.read

Account in use · read-only

No parameters.

download_filesDownload chosen files into --download-dir; skips files already there.read

Download files

Parameter	Type	Bounds	Description
`identifier` required	text
`files`	list of text	max items 200	As get_item lists them.
`formats`	list of text	max items 50	As get_item lists them.
`glob`	text	max length 500	Names, e.g. '.mp3\|.txt'.
`exclude`	text	max length 500	Names to leave out.

get_file_linksDownload URLs, sizes and md5s of chosen files (default: all).read

File download links · read-only

Parameter	Type	Bounds	Description
`identifier` required	text
`files`	list of text	max items 200	As get_item lists them.
`formats`	list of text	max items 50	As get_item lists them.
`glob`	text	max length 500	Names, e.g. '.mp3\|.txt'.
`exclude`	text	max length 500	Names to leave out.
`offset`	integer	min 0
`limit`	integer	min 1; max 500

get_full_textA book's OCR text in chunks, from a byte offset or a scan page.read

Read a book's OCR text · read-only

Parameter	Type	Bounds	Description
`identifier` required	text
`offset`	integer	min 0	In bytes: the last next_offset.
`scan_page`	integer	min 1	Start at this scan page instead.
`max_chars`	integer	min 1000; max 50000
`file`	text	max length 500	A .txt file to read instead.
`book_prefix`	text		File prefix of one book in a multi-book item; default the first.

get_hocrOne scan page's OCR text and words, with pixel boxes and confidence.read

OCR words on a book page · read-only

Parameter	Type	Bounds	Description
`identifier` required	text
`scan_page` required	integer	min 1	As in search_inside and page images.
`words_limit`	integer	min 0; max 3000
`book_prefix`	text		File prefix of one book in a multi-book item; default the first.

get_itemAn item's metadata, files (name, format, size, md5) and links.read

Show an item · read-only

Parameter	Type	Bounds	Description
`identifier` required	text
`files_offset`	integer	min 0
`files_limit`	integer	min 0; max 500	0: none.
`file_source`	one of all, original, derivative, metadata
`max_field_chars`	integer	min 100; max 20000	Cut longer values.

get_reviewsAn item's reviews, newest first, and the average rating.read

Read an item's reviews · read-only

Parameter	Type	Bounds	Description
`identifier` required	text
`mine`	yes/no		Only your own review (needs keys).
`offset`	integer	min 0
`limit`	integer	min 1; max 50
`max_body_chars`	integer	min 100; max 10000	Cut longer reviews.

get_simplelistsThe Simple Lists an item is in, or the members of its own lists.read

Show an item's simple lists · read-only

Parameter	Type	Bounds	Description
`identifier` required	text
`members`	yes/no		List the members of its own lists instead.
`list_name`	text		With members: only this list, e.g. 'holdings'.
`limit`	integer	min 1; max 100
`page`	integer	min 1; max 1000

get_subtitlesAn item's subtitle files, or one read as timestamped text.read

Subtitles · read-only

Parameter	Type	Bounds	Description
`identifier` required	text
`file`	text	max length 500	To read; omit to list them.
`start`	integer	min 0; max 864000	Seconds in.
`offset`	integer	min 0	The last next_offset.
`max_chars`	integer	min 1000; max 50000
`glob`	text	max length 500	Names, e.g. '.mp3\|.txt'.

get_tasksAn item's queued, running and finished tasks. Needs keys.read

List an item's tasks · read-only

Parameter	Type	Bounds	Description
`identifier` required	text
`which`	one of all, active, history		active (queued, running), history (finished) or all.
`limit`	integer	min 1; max 100
`cursor`	text	max length 500	The last next_cursor.

list_collectionOne page of a collection's items, with the total.read

List a collection's items · read-only

Parameter	Type	Bounds	Description
`identifier` required	text
`mediatype`	text		e.g. 'texts'.
`fields`	list of text	max items 50	Default: identifier, title, mediatype, creator, date, downloads.
`sort`	list of text	max items 5	Default: ['downloads desc'].
`limit`	integer	min 1; max 100
`page`	integer	min 1; max 1000

my_itemsYour uploads, newest first: done or still processing. Needs keys.read

List your uploads · read-only

Parameter	Type	Bounds	Description
`pending_only`	yes/no		Only items still processing or failed.
`limit`	integer	min 1; max 100
`page`	integer	min 1; max 1000

search_full_textTexts whose OCR text matches, with the matching passages.read

Full-text search · read-only

Parameter	Type	Bounds	Description
`query` required	text	max length 1000	Words or a "phrase"; Lucene syntax.
`limit`	integer	min 1; max 50
`page`	integer	min 1; max 1000

search_insideFind a word or phrase in a scanned book: scan pages and context.read

Search inside a book · read-only

Parameter	Type	Bounds	Description
`identifier` required	text
`query` required	text	max length 200
`limit`	integer	min 1; max 100
`book_prefix`	text		File prefix of one book in a multi-book item; default the first.

search_itemsOne page of matching items, and the total.read

Search archive.org items · read-only

Parameter	Type	Bounds	Description
`query` required	text	max length 2000	e.g. 'title:(moby dick) AND mediatype:texts'.
`fields`	list of text	max items 50	Default: identifier, title, mediatype, creator, date, collection.
`sort`	list of text	max items 5	e.g. ['downloads desc'].
`limit`	integer	min 1; max 100
`page`	integer	min 1; max 1000

search_items_deepsearch_items for large result sets, paged by cursor.read

Page through big searches · read-only

Parameter	Type	Bounds	Description
`query` required	text	max length 2000
`fields`	list of text	max items 20
`sort`	list of text	max items 5	Put 'identifier' last.
`limit`	integer	min 100; max 500
`cursor`	text	max length 2000	The last next_cursor.

add_reviewPost your review of an item, replacing your earlier one.write

Review an item · destructive

Parameter	Type	Bounds	Description
`identifier` required	text
`title` required	text	max length 255	One line.
`body` required	text	max length 20000
`stars`	integer	min 0; max 5	0 to 5; leave out for no rating.
`confirmation_id`	text	max length 64	Omit to preview; then the preview's id.

copy_filesCopy chosen files into another item, or the same one.write

Copy files · destructive

Parameter	Type	Bounds	Description
`identifier` required	text
`to_identifier` required	text
`files`	list of text	max items 200	As get_item lists them.
`formats`	list of text	max items 50	As get_item lists them.
`glob`	text	max length 500	Names, e.g. '.mp3\|.txt'.
`exclude`	text	max length 500	Names to leave out.
`to_folder`	text	max length 200	A folder in the destination.
`to_name`	text	max length 512	New name (one file only).
`derive`	yes/no		Derive the destination afterwards.
`confirmation_id`	text	max length 64	Omit to preview; then the preview's id.

flag_itemAdd or remove your flag on an item (for the Archive's staff).write

Flag an item · destructive

Parameter	Type	Bounds	Description
`identifier` required	text
`category` required	text
`action`	one of add, remove
`confirmation_id`	text	max length 64	Omit to preview; then the preview's id.

job_cancelStop a job; finished files stay.write

Cancel a job

Parameter	Type	Bounds	Description
`job_id` required	text		The job_id a write tool returned.

job_statusProgress and results of a background job, or a list of your jobs.write

Show a job · read-only

Parameter	Type	Bounds	Description
`job_id`	text		Omit to list your jobs.
`wait_seconds`	integer	min 0; max 300	Wait for the job to end, sending progress.

modify_metadataSet, append or remove metadata of an item or its files; previews a diff first.write

Change metadata · destructive

Parameter	Type	Bounds	Description
`identifier` required	text
`set_fields`	object		Field to value(s), e.g. {'subject': ['maps']}.
`append_values`	object
`remove_values`	object
`remove_fields`	list of text	max items 50	Fields to remove entirely.
`files`	list of text	max items 50	Change these files' metadata instead of the item's.
`confirmation_id`	text	max length 64	Omit to preview; then the preview's id.

move_filesMove chosen files to another item, or rename them within one.write

Move files · destructive

Parameter	Type	Bounds	Description
`identifier` required	text
`to_identifier` required	text
`files`	list of text	max items 200	As get_item lists them.
`formats`	list of text	max items 50	As get_item lists them.
`glob`	text	max length 500	Names, e.g. '.mp3\|.txt'.
`exclude`	text	max length 500	Names to leave out.
`to_folder`	text	max length 200	A folder in the destination.
`to_name`	text	max length 512	New name (one file only).
`derive`	yes/no		Derive the destination afterwards.
`confirmation_id`	text	max length 64	Omit to preview; then the preview's id.

submit_taskRerun derivation on an item you uploaded (e.g. after uploads).write

Rerun a derive · destructive

Parameter	Type	Bounds	Description
`identifier` required	text
`task`	text		Only 'derive' (remake OCR, PDFs, thumbnails).
`remove_derived`	text	max length 200	Glob of derived files to delete first, e.g. '*.pdf'.
`confirmation_id`	text	max length 64	Omit to preview; then the preview's id.

update_simplelistAdd an item to a collection's simple list, or remove it.write

Change a simple list · destructive

Parameter	Type	Bounds	Description
`identifier` required	text
`parent` required	text		The list's parent collection.
`list_name` required	text		The list, e.g. 'holdings'.
`action`	one of add, remove
`notes`	text	max length 2000	With add.
`confirmation_id`	text	max length 64	Omit to preview; then the preview's id.

upload_filesCreate an item or add files to one, from --upload-root folders, as a job.write

Upload files · destructive

Parameter	Type	Bounds	Description
`identifier` required	text
`paths` required	list of text	max items 100	Files or folders in an --upload-root.
`rights_confirmed`	yes/no		True only if the user says they hold the rights.
`title`	text	max length 500	A new item's title.
`mediatype`	one of texts, movies, audio, image, data, software		A new item's mediatype.
`collection`	text		Default: the mediatype's community one. test_collection: trials.
`metadata`	object		A new item's other fields (creator, date, ...).
`to_folder`	text	max length 200	A folder in the item.
`derive`	yes/no		Derive afterwards (OCR, previews).
`confirmation_id`	text	max length 64	Omit to preview; then the preview's id.

delete_filesDelete chosen files (or all_files) of an item; the preview lists each one.delete

Delete files · destructive

Parameter	Type	Bounds	Description
`identifier` required	text
`files`	list of text	max items 200	As get_item lists them.
`formats`	list of text	max items 50	As get_item lists them.
`glob`	text	max length 500	Names, e.g. '.mp3\|.txt'.
`exclude`	text	max length 500	Names to leave out.
`all_files`	yes/no		Every file of the item, instead of a choice.
`include_derivatives`	yes/no		Also their derived files.
`keep_old_versions`	yes/no		Keep copies in history/files.
`confirmation_id`	text	max length 64	Omit to preview; then the preview's id.

delete_reviewDelete your own review of an item; the preview shows it first.delete

Delete your review · destructive

Parameter	Type	Bounds	Description
`identifier` required	text
`confirmation_id`	text	max length 64	Omit to preview; then the preview's id.

8 tools

openlibrary

Open Library: books, editions, authors, subjects, covers and scans.

book_availabilityCan a book be read or borrowed now (1-hour, 14-day loans)? With the borrow link.read

Book availability · read-only

Parameter	Type	Bounds	Description
`identifier`	text	max length 100	An archive.org book.
`book_id`	text	max length 300	Or an Open Library edition or work ID, or ISBN.

openlibrary_authorAn author: names, dates, biography, photo, best-known works.read

Open Library author · read-only

Parameter	Type	Bounds	Description
`author` required	text	max length 300	Author ID like 'OL34184A', or its URL.
`works_limit`	integer	min 0; max 50

openlibrary_cover_urlBuild a covers.openlibrary.org image URL (downloads nothing).read

Open Library cover URL · read-only

Parameter	Type	Bounds	Description
`by` required	one of id, olid, isbn, oclc, lccn		What value is.
`value` required	text	max length 40
`kind`	one of book, author		Book cover or author photo.
`size`	one of S, M, L

openlibrary_editionOne edition: publisher, date, ISBNs, pages, contents, its scan.read

Open Library edition · read-only

Parameter	Type	Bounds	Description
`edition` required	text	max length 300	Edition ID like 'OL7353617M', an ISBN, or a URL.

openlibrary_scanThe archive.org scans of an edition or work, and whether each is public.read

Scans of a book · read-only

Parameter	Type	Bounds	Description
`book_id` required	text	max length 300	Edition ID, ISBN, or work ID (for scans of any edition).
`limit`	integer	min 1; max 20
`check_access`	yes/no		Check each scan is public (a request each).

openlibrary_searchSearch Open Library books (works); 'ia' lists archive.org scans.read

Search Open Library books · read-only

Parameter	Type	Bounds	Description
`query`	text	max length 500	Words or Open Library syntax, e.g. 'subject:whaling'.
`title`	text	max length 300
`author`	text	max length 200
`fields`	list of text	max items 30	Default: key, title, author_name, first_publish_year, ia and more.
`sort`	one of relevance, new, old, editions, rating, readinglog, title, random		'editions': most-published first.
`lang`	text		Prefer editions in this language, e.g. 'fr'.
`limit`	integer	min 1; max 100
`page`	integer	min 1; max 100

openlibrary_subjectWorks on a subject, place, person or time, with the total.read

Open Library subject · read-only

Parameter	Type	Bounds	Description
`subject` required	text	max length 100	e.g. 'whaling', or 'place:...', 'person:...', 'time:...'.
`limit`	integer	min 1; max 50
`offset`	integer	min 0; max 10000
`ebooks_only`	yes/no		Only works with a scan on archive.org.

openlibrary_workA work (a book across editions): authors, description, subjects, covers.read

Open Library work · read-only

Parameter	Type	Bounds	Description
`work` required	text	max length 300	Work ID like 'OL45804W', or its URL.
`editions_limit`	integer	min 0; max 50

2 tools

tvnews

The TV News Archive: program search and closed captions.

tvnews_captionsUp to 5 minutes of a program's captions, with times and a clip link.read

TV News captions · read-only

Parameter	Type	Bounds	Description
`identifier` required	text	max length 100	From tvnews_search.
`start`	integer	min 0; max 86400	Seconds into the program.
`duration`	integer	min 10; max 300	Seconds to read.
`around`	integer	min 0; max 86400	Centre the window here instead (seconds in).

tvnews_searchFind TV news programs by words, channel, program and air date.read

Search TV News · read-only

Parameter	Type	Bounds	Description
`query`	text	max length 500	In titles, descriptions, topics; not captions.
`channel`	text		e.g. CNNW, FOXNEWSW, BBCNEWS.
`program`	text	max length 200
`from_date`	text
`to_date`	text
`sort`	one of newest, oldest, relevance
`limit`	integer	min 1; max 50
`page`	integer	min 1; max 100

1 tool

views

View counts for items and collections.

item_viewsView counts (all time, last 30 and 7 days) for items or collections.read

Item view counts · read-only

Parameter	Type	Bounds	Description
`identifiers` required	list of text	max items 50	Items or collections (which count their items' views).
`days`	integer	min 0; max 90	Also daily views for the last N days (at most 10 identifiers).

6 tools

wayback

The Wayback Machine: availability, CDX capture history, page text and diffs.

wayback_availableThe Wayback capture of a page closest to a time, if any.read

Closest Wayback capture · read-only

Parameter	Type	Bounds	Description
`url` required	text	max length 2048	http or https.
`timestamp`	text		Closest to this time. Default: the newest.

wayback_cdxA page's Wayback captures (CDX). Syntax: internet-archive://wayback/syntax.read

Wayback captures (CDX) · read-only

Parameter	Type	Bounds	Description
`url` required	text	max length 2048	Page address; with match_type, where to match from.
`match_type`	one of exact, prefix, host, domain		prefix: pages under the path; domain: with subdomains.
`from_timestamp`	text		Earliest, e.g. '2015'.
`to_timestamp`	text		Latest, e.g. '20201231'.
`filters`	list of text	max items 5	'field:regex', '!' first excludes, e.g. 'statuscode:200'.
`collapse`	list of text	max items 3	Skip rows repeating a field: 'digest' (content changes), 'timestamp:8' (daily).
`fields`	list of one of urlkey, timestamp, original, mimetype, statuscode, digest, length	max items 7	Columns. Default: all but urlkey.
`limit`	integer	min 1; max 500
`newest`	yes/no		The newest captures instead of the oldest (no cursor).
`cursor`	text		The last next_cursor.

wayback_compareWhat changed between two captures of a page, as a diff of their text.read

Compare two Wayback captures · read-only

Parameter	Type	Bounds	Description
`url` required	text	max length 2048	http or https.
`from_timestamp` required	text		The earlier capture.
`to_timestamp`	text		The later one. Default: the newest.
`context_lines`	integer	min 0; max 10	Unchanged lines around changes.
`max_lines`	integer	min 10; max 2000

wayback_fetchA capture of a page as clean text, in chunks.read

Read a Wayback capture · read-only

Parameter	Type	Bounds	Description
`url` required	text	max length 2048	http or https.
`timestamp`	text		Closest to this time. Default: the newest.
`offset`	integer	min 0; max 1000000
`max_chars`	integer	min 500; max 50000
`include_title`	yes/no
`include_links`	yes/no		Up to 200.

wayback_save_statusThe status of a Save Page Now capture job. Needs keys.read

Save Page Now status · read-only

Parameter	Type	Bounds	Description
`job_id` required	text		The job_id Save Page Now returned.

wayback_saveAsk the Wayback Machine to capture one page now.write

Save Page Now

Parameter	Type	Bounds	Description
`url` required	text	max length 2048	http or https.
`capture_all`	yes/no		Capture error pages (4xx/5xx) too.
`capture_screenshot`	yes/no		Add a screenshot.
`skip_first_archive`	yes/no		Skip the first-capture check (faster).
`if_not_archived_within`	text		Skip if a capture is newer, e.g. '3d 5h' or '3600'.
`delay_wb_availability`	yes/no		Publish after ~12 hours (eases load).
`confirmation_id`	text	max length 64	Omit to preview; then the preview's id.