Anime Release & Legal Streaming Discovery Intelligence

Resolve anime titles into normalized metadata, release schedules, source/adaptation relations, provider-level streaming discovery, and optional raw legal-link evidence. Each dataset row represents one distinct, high-confidence anime entity.

This Actor is discovery intelligence for catalogs, research, and monitoring pipelines—not a generic AniList dump.

Store short description: resolve anime titles into normalized metadata, schedules, source relations, and legal streaming discovery links with conservative matching.

What this Actor does

• Resolves English, Romaji, native, exact, short, and broad anime queries.
• Keeps TV series, seasons, movies, specials, OVAs, ONAs, and remakes separate when they are distinct entities.
• Normalizes identity, external IDs, release dates, airing state, studios, relations, and source provenance.
• Aggregates streaming evidence to one provider row while retaining raw links when requested.
• Uses AniList through a replaceable, operator-gated adapter and Jikan as a replaceable fallback.
• Keeps metadata usable when an optional source or streaming discovery step fails.

Use cases

• Anime catalog enrichment
• Current-airing and upcoming-release tracking
• Scheduled release monitoring
• Legal streaming-link discovery
• Adaptation and source-material mapping
• Anime databases, apps, content, and research pipelines

Quick start

{
  "query": "Cowboy Bebop",
  "maxResults": 1,
  "includeSchedule": true,
  "includeStreamingLinks": true,
  "includeRelations": true,
  "includeAdult": false
}

The default is Store-safe: one result, schedule and streaming discovery enabled, relations enabled, and adult results disabled. See samples/input.default.json.

Input semantics

query accepts an English, Romaji, or native title and is required. maxResults is the maximum number of distinct high-confidence anime entities returned. It is not a request to fill the dataset with weak or ambiguous matches. Exact title queries commonly return one row; broad queries can return several. Duplicate AniList/MAL entities are removed.

includeSchedule, includeStreamingLinks, includeRelations, and includeAdult control optional output and source work. When includeStreamingLinks is false, both streamingDiscovery and streamingProviders are omitted. Source and commercial-policy configuration is operator-side, not consumer input.

Main output

Each row follows this shape:

{
  "recordType": "anime_release_legal_streaming_discovery",
  "query": "Cowboy Bebop",
  "entityMatch": { "confidence": 1, "level": "high", "ambiguity": false },
  "identity": {
    "canonicalTitle": "Cowboy Bebop",
    "format": "TV",
    "episodes": 26,
    "season": "SPRING",
    "seasonYear": 1998,
    "sourceMaterial": "ORIGINAL"
  },
  "externalIds": { "mal": 1, "anilist": 1 },
  "release": {
    "status": "FINISHED",
    "startDate": "1998-04-03",
    "endDate": "1999-04-24",
    "currentlyAiring": false,
    "nextEpisode": null,
    "nextAiringAt": null,
    "scheduleStatus": "unknown",
    "source": "anilist"
  },
  "studios": ["Sunrise"],
  "streamingProviders": [
    {
      "provider": "crunchyroll",
      "source": "anilist",
      "linkType": "episode_links",
      "linkCount": 26,
      "confidence": 0.45,
      "sampleUrl": "https://www.crunchyroll.com/watch/example"
    }
  ],
  "streamingDiscovery": [
    {
      "provider": "crunchyroll",
      "type": "legal_streaming_link",
      "url": "https://www.crunchyroll.com/watch/example",
      "source": "anilist",
      "confidence": 0.45,
      "evidence": ["streamingEpisodes"]
    }
  ],
  "relations": [],
  "dataQuality": {
    "entityMatch": { "score": 1, "level": "high" },
    "metadata": { "score": 0.9, "level": "high" },
    "schedule": { "score": 0.55, "level": "low" },
    "streamingDiscovery": { "score": 0.45, "level": "low" },
    "overall": { "score": 0.76, "level": "medium" },
    "status": "complete",
    "warnings": []
  },
  "sources": [{ "name": "anilist", "status": "success", "warnings": [], "errorCode": null }],
  "fetchedAt": "2026-08-08T00:00:00.000Z"
}

See the complete example in samples/output.default.json.

Output fields

Field	Meaning
identity.canonicalTitle	Canonical title selected for this distinct anime entity.
identity.format	Format such as TV, MOVIE, SPECIAL, OVA, or ONA.
release.status	Authoritative source release status.
release.seasonYear	Release season year when supplied by the source.
identity.episodes	Episode count when supplied by the source.
release.currentlyAiring	Source-backed airing indicator.
release.nextEpisode	Next episode number when schedule data provides it.
release.nextAiringAt	Next airing timestamp when schedule data provides it.
streamingProviders	One provider-level aggregate per normalized provider when streaming discovery is enabled.
identity.sourceMaterial	Source-material classification when supplied by the source.
dataQuality.entityMatch	Entity-match evidence score and level.
dataQuality.overall	Overall evidence-quality score and level.
entityMatch	Public match confidence and ambiguity status.
identity	Canonical titles, format, episodes, season/year, and source material.
externalIds	AniList and MAL identifiers when supplied.
release	Dates, status, airing state, and schedule fields.
studios	Normalized studio names when supplied.
streamingDiscovery	Deduplicated raw legal-link evidence when enabled.
relations	One-level normalized related-anime records.
dataQuality	Component evidence-quality scores, overall status, and warnings.
sources	Concise source outcome and warning records.
query	Original normalized search query.
fetchedAt	Observation timestamp.

Entity matching and release semantics

Matching is conservative. Exact title conflicts remain ambiguous, low-confidence candidates are skipped, and the Actor never selects a false entity merely to satisfy maxResults. Distinct seasons, films, specials, and related formats remain separate when they genuinely match.

release.status is the authoritative status field. startDate, endDate, season, and seasonYear are normalized source values. currentlyAiring, nextEpisode, and nextAiringAt are populated only from source schedule data; the Actor does not infer missing schedule values. scheduleStatus is available, unknown, or not_requested.

Relations are one level deep and deterministic. Common relationType values include SOURCE, ADAPTATION, PREQUEL, and SEQUEL; each relation contains mediaType, a title, and nullable AniList/MAL IDs. Relation trees are not recursively fetched.

Streaming discovery semantics

When enabled, streamingDiscovery contains deduplicated raw legal-link evidence and streamingProviders contains one normalized aggregate per provider:

• linkCount counts valid unique links for that provider.
• sampleUrl is one representative link.
• source preserves provenance and confidence describes evidence quality.
• Provider links are discovery evidence only. They are not availability, entitlement, licensing, country, or playback claims.

The Actor does not implement regional availability, regional proxy logic, TMDB, JustWatch, Crunchyroll scraping, or piracy-link collection.

Confidence and data quality

Scores represent evidence quality, not statistical probabilities. entityMatch, metadata, schedule, streamingDiscovery, and overall each expose a score and low/medium/high level. dataQuality.status is complete, partial, or metadata_only; warnings are concise and actionable. Source request counts, latency, and fallback diagnostics are kept in RUN_SUMMARY.

API and scheduling

Use the Apify API or client libraries to start the Actor with the input above and read the default dataset:

apify call dhHckfKSDz9g7Dnkc \
  --input-file samples/input.default.json \
  --json

For recurring monitoring, schedule the Actor with includeSchedule: true and compare release.nextEpisode, release.nextAiringAt, and fetchedAt between runs. Keep maxResults small for exact titles and increase it only for broad discovery queries. Every successful normalized Dataset row is one billable anime-result event.

Pricing

This Actor uses pay-per-event pricing at $0.010 per successfully produced normalized anime result. Examples are approximately $0.01 for one exact-title result, $0.03 for three broad-query results, up to $0.10 for ten results, and up to $0.20 for twenty results. Actual charges depend on the number of distinct high-confidence entities written to the Dataset; candidates, raw links, provider aggregates, source requests, and ambiguous or zero-result queries are not billable events.

The Actor is published on Apify Store. Pricing is pay-per-event, and users can set a maximum run charge.

Source policy and limitations

AniList is enabled only through the replaceable adapter and an explicit operator policy gate. Use ANILIST_COMMERCIAL_POLICY=under_threshold only while the operator has confirmed monthly revenue remains below $150; use licensed after a commercial license is confirmed. The Actor cannot reliably know monthly revenue and does not automate enforcement. Do not mass-collect, hoard, or use AniList as backup storage.

Jikan is a fallback adapter and is not assumed reliable. Source failures can produce partial output or no result, while a usable source result remains non-fatal. Respect upstream terms, rate limits, attribution requirements, and freshness limits.

Streaming links are discovery evidence only. They do not prove availability, playback entitlement, or licensing status for a viewer. Jikan remains fallback-only and is not assumed reliable.

Responsible use

Do not use discovered links to bypass access controls or geo restrictions. Validate freshness and source permissions before using results in paid or user-facing products.