anidump API

A small, read-only HTTP mirror of AniDB, MyAnimeList, AniList, Anime News Network and Kitsu anime metadata — including Kitsu's per-episode records.

anidump collects public anime metadata from AniDB, MyAnimeList, AniList, Anime News Network and Kitsu and serves it back as plain JSON (and, for AniDB, the original XML). It is a personal hobby project, not an official AniDB, MyAnimeList, AniList, Anime News Network or Kitsu service. Everything here is read-only and unauthenticated — there is no API key, no login, and no write endpoint.

Best-effort dataset: no uptime, latency, or rate-limit guarantees, and the shape below can change. If you need reliability or bulk access, mirror the data rather than hammering this endpoint.

Base URL: https://anidump.id

Data model, in one breath

There are five independent datasets, each keyed by its own upstream id: AniDB records under /api/anidb/…, keyed by an AniDB aid (anidb_id); MyAnimeList records under /api/mal/…, keyed by a mal_id; AniList records under /api/anilist/…, keyed by an anilist_id; Anime News Network encyclopedia records under /api/ann/…, keyed by an ann_id; and Kitsu records under /api/kitsu/…, keyed by a kitsu_id. The datasets are not cross-linked here, with three convenience exceptions: an AniList record carries id_mal, AniList's own pointer to the matching mal_id; an AniDB record carries mal_id and ann_id; and a Kitsu record carries mal_id, anilist_id and anidb_id. (If you want the links themselves rather than one record's view of them, use /api/map.) anidump fetches records one at a time, so only a subset of each upstream's id space is present at any moment, and that subset grows over time. Each dataset has two levels of detail:

Parsed — a flat, curated projection (/api/anidb/{id}, /api/mal/{id}, /api/anilist/{id}, /api/ann/{id}, /api/kitsu/{id}).
Raw — exactly what the upstream returned, byte-for-byte (/raw). AniDB additionally serves its original XML at /raw/xml; MyAnimeList, AniList and Kitsu serve JSON only, and ANN's XML is converted to JSON when we fetch it, so there is no /api/mal/{id}/raw/xml, /api/anilist/{id}/raw/xml, /api/ann/{id}/raw/xml or /api/kitsu/{id}/raw/xml.

Kitsu has one thing the other four do not: episodes. They live nested under the anime (/api/kitsu/{id}/episodes), because that is what they are — an episode belongs to a series, and we fetch a whole anime's episodes as a single unit. There is no flat /api/kitsu/episodes/{id} and no all-episodes listing; walk /api/kitsu/ids and ask each anime for its own. Everywhere else, an "episode" is just a count on the anime record.

Routes

GET/api/anidb/ids

Every aid anidump has actually fetched, with its main title, ordered by anidb_id. This is a subset of AniDB's full aid space and grows over time. Every id it returns resolves on /api/anidb/{id}. An empty dataset returns [], never null. title may be null.

curl https://anidump.id/api/anidb/ids

[
                    {"anidb_id": 1, "title": "Seikai no Monshou"},
                    {"anidb_id": 2, "title": "Seikai no Senki"},
                    {"anidb_id": 5, "title": "Cowboy Bebop: Tengoku no Tobira"}
                    ]

GET/api/anidb/{id}

The parsed record for one aid. 404 if that aid has not been fetched. See the field reference below.

curl https://anidump.id/api/anidb/1

{
                    "anidb_id": 1,
                    "title": "Seikai no Monshou",
                    "title_english": "Crest of the Stars",
                    "title_native": "星界の紋章",
                    "synonyms": ["CotS", "Hvězdný erb"],
                    "type": "TV Series",
                    "restricted": false,
                    "picture": "224210.jpg",
                    "url": "http://www.sunrise-inc.co.jp/seikai/",
                    "startdate": "1999-01-03",
                    "startyear": 1999,
                    "enddate": "1999-03-28",
                    "episodecount": 13,
                    "description": "In the future, mankind ...",
                    "mal_id": 290,
                    "ann_id": 14,
                    "tags": [{"anidb_tag_id": 2626, "name": "future", "weight": 600}],
                    "relatedanime": [{"anidb_id": 2, "type": "Sequel"}]
                    }

GET/api/anidb/{id}/raw

The AniDB HTTP-API response for this aid converted to JSON, served verbatim as stored. Richer and more deeply nested than the parsed record (attributes become @-prefixed keys). 404 if the aid is unknown or its payload has not been fetched yet.

curl https://anidump.id/api/anidb/1/raw

{"anime": {
                    "@id": "1",
                    "@restricted": "false",
                    "type": "TV Series",
                    "episodecount": "13",
                    "startdate": "1999-01-03",
                    "enddate": "1999-03-28",
                    "titles": {"title": [
                    {"@xml:lang": "x-jat", "@type": "main", "#text": "Seikai no Monshou"}
                    ]},
                    "picture": "224210.jpg"
                    }}

GET/api/anidb/{id}/raw/xml

The original AniDB XML, served verbatim as application/xml. 404 if the aid is unknown or the XML has not been fetched yet.

curl https://anidump.id/api/anidb/1/raw/xml

<?xml version="1.0" encoding="UTF-8"?>
                    <anime id="1" restricted="false">
                    <type>TV Series</type>
                    <episodecount>13</episodecount>
                    <startdate>1999-01-03</startdate>
                    <enddate>1999-03-28</enddate>
                    <titles>
                    <title xml:lang="x-jat" type="main">Seikai no Monshou</title>
                    </titles>
                    <picture>224210.jpg</picture>
                    </anime>

GET/api/mal/ids

Every mal_id anidump has actually fetched, with its main title, ordered by mal_id. Every id it returns resolves on /api/mal/{id}. An empty dataset returns [], never null. title may be null.

curl https://anidump.id/api/mal/ids

[
                    {"mal_id": 1, "title": "Cowboy Bebop"},
                    {"mal_id": 5, "title": "Cowboy Bebop: Tengoku no Tobira"},
                    {"mal_id": 6, "title": "Trigun"}
                    ]

GET/api/mal/{id}

The parsed record for one mal_id. 404 if that id has not been fetched. Scores, ranks and member counts are not in this projection — they change constantly, so they are served only by /api/mal/{id}/raw. See the field reference below.

curl https://anidump.id/api/mal/1

{
                    "mal_id": 1,
                    "title": "Cowboy Bebop",
                    "title_english": "Cowboy Bebop",
                    "title_japanese": "カウボーイビバップ",
                    "title_synonyms": [],
                    "type": "TV",
                    "source": "Original",
                    "status": "Finished Airing",
                    "rating": "R - 17+ (violence & profanity)",
                    "approved": true,
                    "airing": false,
                    "episodes": 26,
                    "duration": "24 min per ep",
                    "aired_from": "1998-04-03",
                    "aired_to": "1999-04-24",
                    "season": "spring",
                    "year": 1998,
                    "genres": ["Action", "Award Winning", "Sci-Fi"],
                    "explicit_genres": [],
                    "themes": ["Adult Cast", "Space"],
                    "demographics": [],
                    "studios": ["Sunrise"],
                    "producers": ["Bandai Visual"],
                    "licensors": ["Funimation", "Bandai Entertainment"],
                    "synopsis": "Crime is timeless…",
                    "url": "https://myanimelist.net/anime/1/Cowboy_Bebop",
                    "image_url": "https://cdn.myanimelist.net/images/anime/4/19644l.jpg",
                    "absent": false,
                    "absent_at": null
                    }

GET/api/mal/{id}/raw

The full upstream MyAnimeList record for this id, served verbatim as stored. Much richer than the parsed record: this is where score, scored_by, rank, popularity, members and favorites live, along with relations, streaming, external, theme and the full images / trailer objects. 404 if the id is unknown or its payload has not been fetched yet.

These counters are a snapshot taken when we last fetched the record, not a live value. Treat them as stale by up to a couple of days.

curl https://anidump.id/api/mal/1/raw

{
                    "mal_id": 1,
                    "url": "https://myanimelist.net/anime/1/Cowboy_Bebop",
                    "title": "Cowboy Bebop",
                    "type": "TV",
                    "episodes": 26,
                    "status": "Finished Airing",
                    "aired": {"from": "1998-04-03T00:00:00+00:00", "to": "1999-04-24T00:00:00+00:00"},
                    "score": 8.75,
                    "scored_by": 1054321,
                    "rank": 46,
                    "popularity": 43,
                    "members": 1980000,
                    "favorites": 82000,
                    "genres": [{"mal_id": 1, "type": "anime", "name": "Action"}],
                    "studios": [{"mal_id": 14, "type": "anime", "name": "Sunrise"}],
                    "relations": [{"relation": "Adaptation", "entry": [{"mal_id": 173, "type": "manga"}]}]
                    }

GET/api/anilist/ids

Every anilist_id anidump has actually fetched, with its romaji title, ordered by anilist_id. Every id it returns resolves on /api/anilist/{id}. An empty dataset returns [], never null. title may be null.

curl https://anidump.id/api/anilist/ids

[
                    {"anilist_id": 1, "title": "Cowboy Bebop"},
                    {"anilist_id": 5, "title": "Cowboy Bebop: Tengoku no Tobira"},
                    {"anilist_id": 6, "title": "Trigun"}
                    ]

GET/api/anilist/{id}

The parsed record for one anilist_id. 404 if that id has not been fetched. Scores, popularity and favourites are not in this projection — they change constantly, so they are served only by /api/anilist/{id}/raw. Air dates are split into _year / _month / _day parts, any of which can be null on its own. See the field reference below.

curl https://anidump.id/api/anilist/1

{
                    "anilist_id": 1,
                    "id_mal": 1,
                    "title_romaji": "Cowboy Bebop",
                    "title_english": "Cowboy Bebop",
                    "title_native": "カウボーイビバップ",
                    "synonyms": ["Kaubōi Bibappu"],
                    "format": "TV",
                    "status": "FINISHED",
                    "source": "ORIGINAL",
                    "country_of_origin": "JP",
                    "is_licensed": true,
                    "episodes": 26,
                    "duration": 24,
                    "start_year": 1998, "start_month": 4, "start_day": 3,
                    "end_year": 1999, "end_month": 4, "end_day": 24,
                    "season": "SPRING",
                    "season_year": 1998,
                    "description": "In the year 2071, humanity has colonized…",
                    "cover_image": "https://s4.anilist.co/file/anilistcdn/media/anime/cover/large/bx1.png",
                    "banner_image": "https://s4.anilist.co/file/anilistcdn/media/anime/banner/1.jpg",
                    "genres": ["Action", "Adventure", "Drama", "Sci-Fi"],
                    "tags": ["Space", "Crime", "Episodic"],
                    "studios": ["Sunrise"],
                    "producers": ["Bandai Visual", "Bandai Entertainment"],
                    "absent": false,
                    "absent_at": null
                    }

GET/api/anilist/{id}/raw

The full upstream AniList GraphQL Media object for this id, served verbatim as stored. Much richer than the parsed record: this is where averageScore, meanScore, popularity and favourites live, along with the full tags (with per-tag rank), relations, trailer, coverImage variants and nextAiringEpisode. 404 if the id is unknown or its payload has not been fetched yet.

These counters are a snapshot taken when we last fetched the record, not a live value. Treat them as stale by up to a day.

curl https://anidump.id/api/anilist/1/raw

{
                    "id": 1,
                    "idMal": 1,
                    "title": {"romaji": "Cowboy Bebop", "english": "Cowboy Bebop", "native": "カウボーイビバップ"},
                    "format": "TV",
                    "status": "FINISHED",
                    "averageScore": 86,
                    "meanScore": 86,
                    "popularity": 456596,
                    "favourites": 32527,
                    "tags": [{"id": 63, "name": "Space", "rank": 94}],
                    "studios": {"nodes": [{"id": 14, "name": "Sunrise", "isAnimationStudio": true}]},
                    "relations": {"edges": [{"relationType": "SIDE_STORY", "node": {"id": 5, "type": "ANIME"}}]},
                    "nextAiringEpisode": null
                    }

GET/api/ann/ids

Every ann_id anidump has actually fetched, with its main title, ordered by ann_id. Every id it returns resolves on /api/ann/{id}. An empty dataset returns [], never null. title may be null.

curl https://anidump.id/api/ann/ids

[
                    {"ann_id": 1, "title": "Angel Links"},
                    {"ann_id": 2, "title": "Apocalypse Zero"},
                    {"ann_id": 4658, "title": "Jinki:Extend"}
                    ]

GET/api/ann/{id}

The parsed record for one ann_id. 404 if that id has not been fetched. ANN's vote-based ratings are not in this projection — they change constantly, so they are served only by /api/ann/{id}/raw. url is the Anime News Network encyclopedia entry for this id; please keep it visible if you display the record (see Attribution at the foot of this page). See the field reference below.

curl https://anidump.id/api/ann/1

{
                    "ann_id": 1,
                    "type": "TV",
                    "precision": "TV",
                    "title": "Angel Links",
                    "alternative_titles": ["Seihō Tenshi Angel Links", "星方天使エンジェルリンクス"],
                    "genres": ["action", "drama", "science fiction"],
                    "themes": ["pirates", "space"],
                    "plot_summary": "Li Meifon is the head of a free protection/security agency…",
                    "episodes": 13,
                    "running_time": "half hour",
                    "vintage": ["1999-04-07 to 1999-06-30"],
                    "start_year": 1999,
                    "objectionable_content": "TA",
                    "official_websites": ["http://www.sunrise-inc.co.jp/seihou/ANGEL-LINKS/story.html"],
                    "picture": "https://cdn.animenewsnetwork.com/thumbnails/max500x600/encyc/A1-51.jpg",
                    "url": "https://www.animenewsnetwork.com/encyclopedia/anime.php?id=1",
                    "absent": false,
                    "absent_at": null
                    }

GET/api/ann/{id}/raw

The full Anime News Network encyclopedia record for this id, served verbatim as stored — ANN's api.xml converted to JSON (attributes become @-prefixed keys, text is #text, repeated siblings become arrays). Much richer than the parsed record: this is where ratings, cast, staff, credit, episode, release and the related-prev / related-next links live. 404 if the id is unknown or its payload has not been fetched yet.

curl https://anidump.id/api/ann/1/raw

{"anime": {
                    "@id": "1",
                    "@type": "TV",
                    "@precision": "TV",
                    "@name": "Angel Links",
                    "info": [
                    {"@type": "Main title", "#text": "Angel Links"},
                    {"@type": "Genres", "#text": "action"},
                    {"@type": "Picture", "@src": "https://cdn.animenewsnetwork.com/…/A1-51.jpg"}
                    ],
                    "ratings": {"@nb_votes": "352", "@bayesian_score": "5.85776", "@weighted_score": "5.8028"},
                    "cast": [{"@lang": "JA", "role": "Li Meifon", "person": {"#text": "Yuko Miyamura"}}]
                    }}

GET/api/kitsu/ids

Every kitsu_id anidump has actually fetched, with its canonical title, ordered by kitsu_id. Every id it returns resolves on /api/kitsu/{id}. An empty dataset returns [], never null. title may be null.

curl https://anidump.id/api/kitsu/ids

[
                    {"kitsu_id": 1, "title": "Cowboy Bebop"},
                    {"kitsu_id": 2, "title": "Cowboy Bebop: Tengoku no Tobira"},
                    {"kitsu_id": 3, "title": "Trigun"}
                    ]

GET/api/kitsu/{id}

The parsed record for one kitsu_id. 404 if that id has not been fetched. Ratings, user counts and the global ranks are not in this projection — they change constantly, so they are served only by /api/kitsu/{id}/raw. url is the Kitsu entry for this id; please keep it visible if you display the record (see Attribution at the foot of this page).

This is the record to reach for if you want cross-service ids: Kitsu keeps its own mapping table, so mal_id, anilist_id and anidb_id come straight from it. See the field reference below.

curl https://anidump.id/api/kitsu/1

{
                    "kitsu_id": 1,
                    "slug": "cowboy-bebop",
                    "canonical_title": "Cowboy Bebop",
                    "title_en": "Cowboy Bebop",
                    "title_en_jp": "Cowboy Bebop",
                    "title_ja_jp": "カウボーイビバップ",
                    "other_titles": ["Cowboy Bebop"],
                    "abbreviated_titles": ["COWBOY BEBOP"],
                    "description": "In the year 2071, humanity has colonized…",
                    "subtype": "TV",
                    "status": "finished",
                    "tba": null,
                    "age_rating": "R",
                    "age_rating_guide": "17+ (Violence & Profanity)",
                    "nsfw": false,
                    "episode_count": 26,
                    "episode_length": 24,
                    "total_length": 624,
                    "start_date": "1998-04-03",
                    "end_date": "1999-04-24",
                    "poster_image": "https://media.kitsu.io/anime/poster_images/1/original.png",
                    "cover_image": "https://media.kitsu.io/anime/cover_images/1/original.png",
                    "youtube_video_id": "qig4KOK2R2g",
                    "url": "https://kitsu.io/anime/cowboy-bebop",
                    "mal_id": 1,
                    "anilist_id": 1,
                    "anidb_id": 23,
                    "absent": false,
                    "absent_at": null
                    }

GET/api/kitsu/{id}/raw

The full upstream Kitsu record for this id, served verbatim as stored: the JSON:API resource object (id, type, attributes) plus a resolved mappings array. Much richer than the parsed record: this is where averageRating, ratingFrequencies, userCount, favoritesCount, popularityRank and ratingRank live, along with every posterImage / coverImage size variant, the whole titles locale map, and nextRelease. 404 if the id is unknown or its payload has not been fetched yet.

Kitsu's JSON:API links and relationships are not kept: they are boilerplate derivable from the id, and relationships in particular is only a pointer into a batch-scoped included block that does not survive one record. The information in it — the cross-service mappings — is resolved at fetch time and stored as the flat mappings array you see here. These counters are a snapshot taken when we last fetched the record, not a live value.

curl https://anidump.id/api/kitsu/1/raw

{
                    "id": "1",
                    "type": "anime",
                    "attributes": {
                    "slug": "cowboy-bebop",
                    "canonicalTitle": "Cowboy Bebop",
                    "titles": {"en": "Cowboy Bebop", "en_jp": "Cowboy Bebop", "ja_jp": "カウボーイビバップ"},
                    "subtype": "TV",
                    "status": "finished",
                    "averageRating": "82.29",
                    "ratingFrequencies": {"2": "13", "3": "1", "4": "6"},
                    "userCount": 138416,
                    "favoritesCount": 3958,
                    "popularityRank": 28,
                    "ratingRank": 27,
                    "posterImage": {"tiny": "…", "small": "…", "original": "…"}
                    },
                    "mappings": [
                    {"id": "1", "externalSite": "myanimelist/anime", "externalId": "1"},
                    {"id": "9578", "externalSite": "anilist/anime", "externalId": "1"},
                    {"id": "43151", "externalSite": "anidb", "externalId": "23"}
                    ]
                    }

GET/api/kitsu/{id}/episodes

Every episode of one anime, in broadcast order (season_number, then number). Kitsu is the only upstream here with a real per-episode dataset, so this is the only place episode titles, air dates and synopses exist — the other four give you an episode count and nothing more.

[] and 404 mean different things, and the difference matters if you are mirroring. An empty episodes array with 200 is a real answer: this anime has no episodes on Kitsu (a movie, an unaired entry, a listing Kitsu has not broken out). A 404 means we have never fetched this anime's episode list — the id may be perfectly valid on /api/kitsu/{id}. Do not record a 404 as "0 episodes".

Episodes are nested under the anime — there is no /api/kitsu/episodes/{id}, and no /api/kitsu/episodes/ids listing. An episode is only meaningful as part of a series, we fetch a whole anime's episodes as one unit, and a flat listing of all ~292,000 of them would be a ~10 MB response nobody wants. Start from /api/kitsu/ids and walk it.

curl https://anidump.id/api/kitsu/1/episodes

{
                    "kitsu_id": 1,
                    "url": "https://kitsu.io/anime/cowboy-bebop",
                    "episodes": [
                    {
                    "kitsu_episode_id": 1,
                    "kitsu_id": 1,
                    "canonical_title": "Asteroid Blues",
                    "title_en_us": "Asteroid Blues",
                    "title_en_jp": "Asteroid Blues",
                    "title_ja_jp": "アステロイドブルース",
                    "other_titles": [],
                    "description": "Spike and Jet pursue a bounty on Tijuana…",
                    "season_number": 1,
                    "number": 1,
                    "relative_number": 1,
                    "airdate": "1998-04-03",
                    "length_minutes": 25,
                    "thumbnail": "https://media.kitsu.io/episodes/thumbnails/1/original.jpg"
                    }
                    ]
                    }

GET/api/kitsu/{id}/episodes/{episode_id}

One episode, exactly as it appears in the list above, plus the anime's url. It is scoped to its anime: an episode_id that belongs to a different anime is a 404 under this path, even though it exists. Both ids must be positive integers, or it is a 400.

curl https://anidump.id/api/kitsu/1/episodes/1

{
                    "kitsu_episode_id": 1,
                    "kitsu_id": 1,
                    "canonical_title": "Asteroid Blues",
                    "title_en_us": "Asteroid Blues",
                    "title_en_jp": "Asteroid Blues",
                    "title_ja_jp": "アステロイドブルース",
                    "other_titles": [],
                    "description": "Spike and Jet pursue a bounty on Tijuana…",
                    "season_number": 1,
                    "number": 1,
                    "relative_number": 1,
                    "airdate": "1998-04-03",
                    "length_minutes": 25,
                    "thumbnail": "https://media.kitsu.io/episodes/thumbnails/1/original.jpg",
                    "url": "https://kitsu.io/anime/cowboy-bebop"
                    }

GET/api/kitsu/{id}/episodes/raw

The whole episode list for this anime as we stored it from Kitsu, verbatim: each entry is the upstream episode object flattened to {id, …attributes}, ordered by episode id. This is where the fields the parsed projection leaves out live — synopsis (a duplicate of description in every episode we measured), createdAt, updatedAt, and the thumbnail object rather than just its original URL. Same 404 rule as above.

curl https://anidump.id/api/kitsu/1/episodes/raw

{
                    "kitsu_id": 1,
                    "episodes": [
                    {
                    "id": "1",
                    "canonicalTitle": "Asteroid Blues",
                    "titles": {"en_us": "Asteroid Blues", "en_jp": "Asteroid Blues", "ja_jp": "アステロイドブルース"},
                    "seasonNumber": 1,
                    "number": 1,
                    "relativeNumber": 1,
                    "airdate": "1998-04-03",
                    "length": 25,
                    "synopsis": "Spike and Jet pursue a bounty on Tijuana…",
                    "description": "Spike and Jet pursue a bounty on Tijuana…",
                    "createdAt": "2013-02-20T16:00:14.183Z",
                    "updatedAt": "2023-11-01T09:12:22.918Z",
                    "thumbnail": {"original": "https://media.kitsu.io/episodes/thumbnails/1/original.jpg"}
                    }
                    ]
                    }

GET/api/map/{id}

The cross-service id map for one MyAnimeList id: the sets of AniDB, AniList and ANN ids that refer to the same title. Each is an array because the services split and merge titles differently, so a MAL entry can correspond to several ids on another service (and a side with none known is []). It is derived first-party — from AniList's idMal, AniDB's <resources> ids, and exact-title matching — so it can contain a link the source itself got wrong, and can miss one the sources never recorded. 404 if the MAL id has no known cross-reference.

curl https://anidump.id/api/map/1535

{"mal_id": 1535, "anidb_id": [4563], "anilist_id": [1535], "ann_id": [6592]}

GET/api/map

The entire map, one object per MAL id (same shape as /api/map/{id}), ordered by mal_id. Like the /ids listings this is a single large response; mirror it rather than polling.

curl https://anidump.id/api/map

[{"mal_id": 1, "anidb_id": [], "anilist_id": [1], "ann_id": []}, …]

Field reference — `/api/anidb/{id}`

Field	Type	Notes
anidb_id	integer	The AniDB aid. Stable primary key.
title	string · null	The AniDB main title (usually romaji / `x-jat`).
title_english	string · null	The official English title, when AniDB has one (`null` for most entries).
title_native	string · null	The official native (Japanese) title. Distinct from `title`, which is the romaji.
synonyms	array of string	Alternative/synonym titles across languages. Empty `[]` when none, never `null`.
type	string · null	e.g. `TV Series`, `Movie`, `OVA`, `Web`.
restricted	boolean	AniDB's adult / 18+ flag. `true` for restricted entries, else `false`.
picture	string · null	A bare cover-art filename, not a URL. Build the image URL yourself: `https://cdn.anidb.net/images/main/<picture>` (e.g. `https://cdn.anidb.net/images/main/224210.jpg`). `null` when the anime has no picture.
url	string · null	The anime's official website, when AniDB records one. `null` otherwise.
startdate	string · null	Full air-start date as `YYYY-MM-DD`. See the note on partial dates below.
startyear	integer · null	The air-start year on its own. Populated even when `startdate` is `null`.
enddate	string · null	Full air-end date as `YYYY-MM-DD`; same partial-date caveat as `startdate`.
episodecount	integer · null	Planned/known episode count. `null` when unknown.
description	string · null	The AniDB synopsis, kept verbatim including AniDB's own markup. `null` when absent.
mal_id	integer · null	The matching MyAnimeList id, taken from AniDB's cross-reference block — join it to `/api/mal/{mal_id}`. `null` when AniDB has no MAL link (~14% of entries).
ann_id	integer · null	The matching Anime News Network encyclopedia id — join it to `/api/ann/{ann_id}`. `null` when AniDB has no ANN link (~32% of entries).
tags	array	AniDB tags, each `{"anidb_tag_id": <int>, "name": "<tag>", "weight": <int\|null>}`, ordered by descending weight (relevance). Empty `[]` or `null` when there are none.
relatedanime	array	Related entries, each `{"anidb_id": <int>, "type": "<relation>"}` where the relation is AniDB's own (`Sequel`, `Prequel`, `Side story`, `Parent story`, …). Empty `[]` or `null` when there are no relations.

startdate vs. startyear. AniDB sometimes knows only a partial date — a bare year, or a year and month — for when a show aired. A partial date cannot be expressed as a real YYYY-MM-DD, so startdate (and enddate) collapse to null in those cases, while startyear keeps the year. So: prefer startdate for a precise date, but fall back to startyear when it is null. The same holds for enddate.

Field reference — `/api/mal/{id}`

Field	Type	Notes
mal_id	integer	The MyAnimeList id. Stable primary key. Unrelated to `anidb_id`.
title	string · null	The default title, usually romaji.
title_english title_japanese	string · null	`null` for a large fraction of entries — most have no English title.
title_synonyms	array of string	Alternative titles. Empty `[]` when there are none, never `null`.
type	string · null	`TV`, `Movie`, `OVA`, `ONA`, `Special`, `TV Special`, `Music`, `CM`, `PV`, or the literal `Unknown`.
source	string · null	What it was adapted from: `Original`, `Manga`, `Light novel`, `Game`, …
status	string · null	`Finished Airing`, `Currently Airing`, or `Not yet aired`.
rating	string · null	Content rating, e.g. `PG-13 - Teens 13 or older`. The literal string `None` means MyAnimeList records no rating.
approved	boolean	Whether MyAnimeList has approved the entry.
airing	boolean	Whether it was airing when we last fetched the record.
episodes	integer · null	`null` when unknown (often for currently-airing or unaired entries).
duration	string · null	Free-form upstream text, not a number: `24 min per ep`, `1 hr 35 min`, `Unknown`. Parse it yourself if you need minutes.
aired_from aired_to	string · null	Air-start / air-end date as `YYYY-MM-DD`. `aired_to` is `null` for well over half of entries (single-episode works, and anything still airing).
season	string · null	`winter`, `spring`, `summer` or `fall`. `null` together with `year` for entries with no seasonal slot.
year	integer · null	The season's year. `null` exactly when `season` is.
genres explicit_genres themes demographics	array of string	Names only (the upstream ids and URLs are dropped; they are in `/raw`). Empty `[]` rather than `null`.
studios producers licensors	array of string	Company names only. Empty `[]` rather than `null`.
synopsis	string · null	Plot summary, as written on MyAnimeList.
url	string · null	Canonical MyAnimeList permalink. It carries a title slug, so it is not derivable from `mal_id` alone.
image_url	string · null	A full cover-art URL on cdn.myanimelist.net (unlike AniDB's `picture`, which is a bare filename). Fetch it from there, not through this API.
absent	boolean	`true` when the entry has disappeared from MyAnimeList's catalogue since we archived it. The record is kept and still served — that is when this mirror is most useful. The flag clears if the entry reappears.
absent_at	string · null	RFC 3339 timestamp of when the entry was first observed missing. `null` whenever `absent` is `false`.

No scores, ranks or member counts here. score, scored_by, rank, popularity, members and favorites move for nearly every title every time we refresh it, so keeping them in the parsed projection would mean rewriting the whole table continuously for numbers that are stale the moment they are served. They are still archived, and /api/mal/{id}/raw returns them verbatim — just read them as a snapshot, not a live figure.

Field reference — `/api/anilist/{id}`

Field	Type	Notes
anilist_id	integer	The AniList id. Stable primary key. Unrelated to `anidb_id` and `mal_id`.
id_mal	integer · null	AniList's own pointer to the matching MyAnimeList id — the one cross-link between datasets, usable against `/api/mal/{id}`. `null` when AniList records no mapping; not guaranteed unique (AniList and MAL split/merge differently).
title_romaji	string · null	AniList's canonical romaji title. Effectively always present.
title_english title_native	string · null	`title_english` is `null` for about half of entries.
synonyms	array of string	Alternative titles. Empty `[]` when there are none, never `null`.
format	string · null	AniList's uppercase enum, verbatim: `TV`, `TV_SHORT`, `MOVIE`, `OVA`, `ONA`, `SPECIAL`, `MUSIC`.
status	string · null	`FINISHED`, `RELEASING`, `NOT_YET_RELEASED`, `CANCELLED`, `HIATUS`.
source	string · null	What it was adapted from: `ORIGINAL`, `MANGA`, `LIGHT_NOVEL`, `VISUAL_NOVEL`, `VIDEO_GAME`, `OTHER`, …
country_of_origin	string · null	ISO 3166 country code: `JP`, `CN`, `KR`, `TW`.
is_licensed	boolean	Whether AniList marks the title as officially licensed.
episodes	integer · null	`null` when unknown (often for currently-airing or unaired entries).
duration	integer · null	Runtime in minutes per episode, as a number (unlike MAL's free-form `duration` string). `null` when unknown.
start_year start_month start_day end_year end_month end_day	integer · null	Air-start / air-end date, as separate parts. AniList often knows only part of a date, so each part is independently `null` — a year with a `null` month and day is common. See the note below.
season	string · null	`WINTER`, `SPRING`, `SUMMER` or `FALL`. `null` together with `season_year` for entries with no seasonal slot.
season_year	integer · null	The premiere season's year. `null` exactly when `season` is.
description	string · null	Synopsis, as written on AniList. Contains HTML markup (`<br>`, `<i>`, …) — render or strip it yourself.
cover_image	string · null	Full cover-art URL on s4.anilist.co (the largest variant). Fetch it from there, not through this API.
banner_image	string · null	Full banner-image URL, or `null` (most entries have none).
genres tags	array of string	Names only. `tags` drops each tag's `id` and relevance `rank` (those are in `/raw`). Empty `[]` rather than `null`.
studios producers	array of string	Company names only. AniList keeps one studio list and flags each with `isAnimationStudio`; `studios` is the animation studios, `producers` the rest. Empty `[]` rather than `null`.
absent	boolean	`true` when the id has disappeared from AniList since we archived it (usually a duplicate AniList merged away). The record is kept and still served — that is when this mirror is most useful. The flag clears if the entry reappears.
absent_at	string · null	RFC 3339 timestamp of when the entry was first observed missing. `null` whenever `absent` is `false`.

AniList dates come in parts. Unlike AniDB and MyAnimeList, which we fold into a single YYYY-MM-DD string, AniList models a date as a year / month / day triple where any part can be missing on its own — a show may have a known year but no month or day. We keep those parts as-is (start_year, start_month, start_day, and the end_* trio) so nothing is lost. Build a full date only when all three parts are present; otherwise use whatever parts you have.

No scores, popularity or favourites here. averageScore, meanScore, popularity and favourites move for nearly every title every time we refresh it, so they are left out of the parsed projection for the same reason as MAL's counters. They are still archived, and /api/anilist/{id}/raw returns them verbatim — read them as a snapshot, not a live figure.

Field reference — `/api/ann/{id}`

Field	Type	Notes
ann_id	integer	The Anime News Network encyclopedia id. Stable primary key. Unrelated to `anidb_id`, `mal_id` and `anilist_id`.
type	string · null	`TV`, `OAV`, `movie`, `ONA`, `special` or `omnibus`.
precision	string · null	ANN's disambiguation label — `TV`, `TV 2`, `movie series`, `meta`, … — which separates entries that share a title. Distinct from `type`.
title	string · null	The ANN main title.
alternative_titles	array of string	Other titles ANN lists (across languages). Empty `[]` rather than `null`.
genres themes	array of string	ANN's genre and theme tags (lowercase). Empty `[]` rather than `null`.
plot_summary	string · null	Plot summary, as written on ANN. `null` when ANN records none.
episodes	integer · null	`null` when unknown, or when ANN writes a non-numeric count (a few entries say e.g. `12 (+ 2 specials)`). The raw text is in `/raw`.
running_time	string · null	Free-form upstream text, not a number: `half hour`, `30 minutes`. Parse it yourself if you need minutes.
vintage	array of string	Release dates/periods, free-form and multi-valued: a full date, a range `A to B`, a bare year, or any of those with a parenthetical region. Empty `[]` rather than `null`. See `start_year` for a filterable year.
start_year	integer · null	Convenience: the leading 4-digit year of the first `vintage`. `null` when it does not begin with a year. Not a substitute for `vintage`.
objectionable_content	string · null	ANN's content code — `TA`, `OC`, `AO`, `AA`, `MA` (upstream casing varies). `null` when ANN records none.
official_websites	array of string	Official site URLs ANN lists. Empty `[]` rather than `null`.
picture	string · null	A full cover-art URL on cdn.animenewsnetwork.com. Fetch it from there, not through this API. `null` when ANN has no picture.
url	string	The Anime News Network encyclopedia entry for this id (animenewsnetwork.com/encyclopedia/anime.php?id=<ann_id>). Always present. Keep it visible when you display the record — see Attribution below.
absent	boolean	`true` when ANN no longer serves this id. The record is kept and still served — that is when this mirror is most useful. The flag clears if the entry reappears.
absent_at	string · null	RFC 3339 timestamp of when the entry was first observed missing. `null` whenever `absent` is `false`.

Anime News Network requires attribution. ANN's terms of use ask that any page showing their encyclopedia details credit Anime News Network as the source and link the relevant entry. Every /api/ann/{id} response carries that link as url; if you display ANN data, surface it. The vote-based ratings (in /raw) are a snapshot from our last fetch, not a live figure.

Field reference — `/api/kitsu/{id}`

Field	Type	Notes
kitsu_id	integer	The Kitsu id. Stable primary key. Unrelated to `anidb_id`, `mal_id`, `anilist_id` and `ann_id`.
slug	string · null	Kitsu's URL slug (`cowboy-bebop`). Not derivable from the id — it is what builds `url`.
canonical_title	string · null	The title Kitsu itself displays. This is what `/api/kitsu/ids` returns as `title`.
title_en title_en_jp title_ja_jp	string · null	The English, romaji and native titles. Kitsu stores titles as an open-ended locale map, not a fixed set of fields, so these three are the common locales pulled out by name; `title_en` is `null` for about half of entries.
other_titles	array of string	Every other locale's title (`en_us`, `zh_cn`, `ko_kr`, …), so nothing in the locale map is lost. Empty `[]` rather than `null`.
abbreviated_titles	array of string	Short forms and alternative spellings Kitsu lists. Empty `[]` rather than `null`.
description	string · null	Synopsis, as written on Kitsu. Plain text (unlike AniList's, which carries HTML).
subtype	string · null	`TV`, `movie`, `ONA`, `OVA`, `special` or `music`. Note the mixed casing — it is Kitsu's, verbatim.
status	string · null	`finished`, `current`, `upcoming`, `unreleased` or `tba`.
tba	string · null	Kitsu's free-text note about an unannounced date — `"2027"`, `"tba"`, a whole sentence. A string, not a boolean, despite the name. `null` for most entries.
age_rating	string · null	`G`, `PG`, `R` or `R18`.
age_rating_guide	string · null	Free-form gloss on the rating, e.g. `17+ (Violence & Profanity)`.
nsfw	boolean	Kitsu's adult flag — the `anidb.restricted` analogue.
episode_count	integer · null	`null` when unknown (often for currently-airing or unaired entries).
episode_length total_length	integer · null	Minutes per episode, and minutes for the whole run — real numbers, unlike MAL's free-form `duration` string and ANN's `running_time`.
start_date end_date	string · null	Air-start / air-end date as `YYYY-MM-DD`. Kitsu emits a full date or nothing, so there is no partial-date caveat here (contrast AniDB's `startyear` and AniList's split date parts).
poster_image cover_image	string · null	Full image URLs on media.kitsu.io, the original size only. The tiny/small/medium/large variants are in `/raw`. Fetch them from there, not through this API.
youtube_video_id	string · null	Bare YouTube id of the trailer, not a URL — build youtube.com/watch?v=<id> yourself.
url	string	The Kitsu entry for this id (kitsu.io/anime/<slug>, falling back to kitsu.io/anime/<kitsu_id> when the slug is `null`). Always present. Keep it visible when you display the record — see Attribution below.
mal_id anilist_id anidb_id	integer · null	Kitsu's own cross-references, taken from its `mappings` — join them to `/api/mal/{id}`, `/api/anilist/{id}` and `/api/anidb/{id}`. Each is `null` when Kitsu records no mapping to that service or more than one: an unambiguous id, or nothing. There is no `ann_id` — Kitsu maps no Anime News Network ids at all. Kitsu's mappings are community-editable, so treat them as a strong hint, not a guarantee.
absent	boolean	`true` when Kitsu no longer serves this id. The record is kept and still served — that is when this mirror is most useful. The flag clears if the entry reappears.
absent_at	string · null	RFC 3339 timestamp of when the entry was first observed missing. `null` whenever `absent` is `false`.

No ratings, user counts or ranks here. averageRating, ratingFrequencies, userCount, favoritesCount, popularityRank and ratingRank are left out of the parsed projection for the same reason as MAL's and AniList's counters — and the two ranks are worse than the rest, being global positions that shift for thousands of titles whenever one of them gains a vote. They are still archived, and /api/kitsu/{id}/raw returns them verbatim — read them as a snapshot, not a live figure.

Field reference — `/api/kitsu/{id}/episodes`

The fields of one episode. The same object appears as an element of /api/kitsu/{id}/episodes and as the whole body of /api/kitsu/{id}/episodes/{episode_id} (which adds url, the anime's Kitsu entry — the list carries it once, at the top level, rather than on every episode).

Field	Type	Notes
kitsu_episode_id	integer	The episode's own Kitsu id — a different id space from `kitsu_id`. Globally unique across all anime, so it is the key, but it only resolves under its own anime's path.
kitsu_id	integer	The anime this episode belongs to — the id in the URL, repeated on every element so a single episode lifted out of the array is still self-describing. Join it to `/api/kitsu/{id}`.
canonical_title	string · null	The title Kitsu itself displays for the episode. Often `null` for never-localised entries, and frequently just `"Episode 12"`.
title_en_us title_en_jp title_ja_jp	string · null	Note the `en_us`. Episodes key their locale map differently from the anime resource: an anime has `en` / `en_jp` / `ja_jp` (hence `title_en` on `/api/kitsu/{id}`), an episode has `en_us` / `en_jp` / `ja_jp`. That is Kitsu's, not ours — an episode has no `en` title at all, so do not expect a `title_en` here.
other_titles	array of string	Every other locale's title, so nothing in the locale map is lost. Empty `[]` rather than `null`.
description	string · null	The episode synopsis. Kitsu also carries a `synopsis` field, but it was identical to this one in every episode we compared, so only this one is projected — both are in `/api/kitsu/{id}/episodes/raw` if you want to check.
season_number	integer · null	Kitsu's own seasoning of the series, not a broadcast season. Most series are all season 1; a few use it to split long runs.
number relative_number	integer · null	The absolute episode number within the series, and the number within its season. Both are nullable and neither is unique — they are community-entered, and duplicates and gaps exist. Sort on them for display (that is the order this route returns), but key on `kitsu_episode_id`.
airdate	string · null	First broadcast, `YYYY-MM-DD`. `null` for unaired episodes and for a great many older ones nobody has filled in.
length_minutes	integer · null	Runtime in minutes (Kitsu's `length`). `null` when unknown.
thumbnail	string · null	Full image URL on media.kitsu.io, the original size only. Fetch it from there, not through this API.
url	string	Only on `/api/kitsu/{id}/episodes/{episode_id}`. The Kitsu entry for the anime (Kitsu has no public per-episode page). On the list route the same link sits once at the top level. Keep it visible when you display the record — see Attribution below.

Errors

Errors are JSON with the matching HTTP status:

{"error": "not found"}

400 — the {id} is not a positive integer.
404 — no such id, its payload has not been fetched, or an unknown path (including /api/mal/{id}/raw/xml, /api/anilist/{id}/raw/xml, /api/ann/{id}/raw/xml and /api/kitsu/{id}/raw/xml, which do not exist).
405 — wrong HTTP method (only GET/HEAD are served).
500 — something broke on our side; the detail is logged, not returned.

anidump API

Data model, in one breath

Routes

GET/api/anidb/ids

GET/api/anidb/{id}

GET/api/anidb/{id}/raw

GET/api/anidb/{id}/raw/xml

GET/api/mal/ids

GET/api/mal/{id}

GET/api/mal/{id}/raw

GET/api/anilist/ids

GET/api/anilist/{id}

GET/api/anilist/{id}/raw

GET/api/ann/ids

GET/api/ann/{id}

GET/api/ann/{id}/raw

GET/api/kitsu/ids

GET/api/kitsu/{id}

GET/api/kitsu/{id}/raw

GET/api/kitsu/{id}/episodes

GET/api/kitsu/{id}/episodes/{episode_id}

GET/api/kitsu/{id}/episodes/raw

GET/api/map/{id}

GET/api/map

Field reference — /api/anidb/{id}

Field reference — /api/mal/{id}

Field reference — /api/anilist/{id}

Field reference — /api/ann/{id}

Field reference — /api/kitsu/{id}

Field reference — /api/kitsu/{id}/episodes

Errors

Attribution

Field reference — `/api/anidb/{id}`

Field reference — `/api/mal/{id}`

Field reference — `/api/anilist/{id}`

Field reference — `/api/ann/{id}`

Field reference — `/api/kitsu/{id}`

Field reference — `/api/kitsu/{id}/episodes`