Give your AI a taste of adult world

The JAV Metadata API for developers & AI agents

Search JAV metadata across multiple providers. Fast, clean, no bullshit.

Give this to your agent

$ set up https://javinfo.dev/SETUP.md

and let it take it from there.

For humans

Get a free key Read the docs

Multiple: Sources
One: Schema

Trusted by 200+ developers

// the whole api, on one screen

Nine reasons your scraper is fired.

One typed endpoint. Every source folded into a single schema. Copy the curl, ship tonight, sleep straight through Tuesday's WAF update.

Read the docs.javinfo.dev for more examples

$ curl -s https://api.javinfo.dev/movie \
     -H "x-javinfo-key: jvi_••••••••" \
     -d '{"q":"SSIS-001"}'

# → 200 OK · 1 result · billed $0.0005
{
  "q":      "SSIS-001",
  "source": "javdb",
  "result": {
    "dvdId":       "SSIS-001",
    "titleEn":     "…",
    "actresses":   ["…"],
    "runtimeMins": 120,
    "extra": {
      "downloadLinks": [{ "magnet": "magnet:?xt=urn:btih:…" }]
    }
  }
}

One key, every source

Every source behind a single request. Your brittle scrapers, retired in one afternoon.

Unblockable by design

They rotate WAFs. We rotate browser fingerprints. Your calls keep landing while the rest eat a 403.

One typed schema

Same shape from every source. OpenAPI spec, RapidAPI-listed, autocompletion out of the box.

Metadata + magnets + streams

Not just titles — download links and HLS masters ship in the same payload. No second call.

Crypto billing

No card. No name. No judgment. Pay in any coin, settle on-chain — billing stays yours.

$0.10 on the house

Free to start, $0.0002 a query after. No plans, no minimums, no sales call. Balance never expires.

curl in 10 seconds

One endpoint, one header, copy-paste, 200 OK. Integration done before your coffee.

The full drop

DVD id, title, actresses + images, gallery, cover art, runtime, genres. All of it, every call.

Built for tools

Jellyfin, Emby, Kodi, Telegram bots. Drop-in metadata that doesn't break every Tuesday.

every source — one key, zero drama.Get in

// 01 — sources

Every source, one schema

javinfo wraps each provider so you integrate once. Add a source and your code doesn't change.

dmm
DMM.com storefront: bilingual metadata with title, cast, maker, label, series, covers, and preview galleries.
fanza
FANZA storefront: the same bilingual metadata, scoped to FANZA's catalog.
javdb
Adds download and magnet links, score, and vote count.
missav
Adds an extra.streams object with HLS playlist URLs (.m3u8) plus per-resolution variants.
javdatabase
Adds description, sample images, and a trailer URL.
javlibrary
Deep-catalog fallback: title, cast, cover, sample images on /movie; keyword /query with num (no filters).

// 02 — use cases

What you can build

Normalized metadata on tap — for the tools you're already building.

Media libraries

Enrich a collection by code: title, cast, covers, runtime and release date — in English and Japanese.

AI agents & tools

Give an LLM a clean, predictable lookup. One schema, no per-source branching, llms.txt for discovery.

Datasets & catalogs

Batch DVD IDs into a normalized dataset. Studio, label, series and cast mapped the same way every time.

Search & discovery

Back a search box with fuzzy code matching — SSIS-001 and ssis001 resolve to the same release.

// 03 — schema

One schema, every source

DMM/FANZA, JavDB, missav or javdatabase — you read result.* the same way. Provider-specific extras ride along under result.extra.*.

Every provider maps to the same base result shape:

dvdId: The DVD ID or code, for example SSIS-001
titleEn / titleJa: Title in English and Japanese
actresses: Cast names as a string array; DMM/FANZA adds profile images under extra.actressesRich
makers / label / series: Studio, label, and series
releaseDate / runtimeMins: Release date and runtime in minutes
jacketFullUrl / galleryFull: Cover art and preview gallery URLs

// 04 — agents

Built for AI agents

API-first, predictable, and honest about what it is: metadata, not media.

Predictable JSON: One envelope, one schema. Safe to parse, easy to prompt against.
llms.txt + discovery: The site serves llms.txt and RFC 8288 discovery headers, so crawlers and agents can find the API.
No auth gymnastics: One key from the app. No OAuth dance, no scraper upkeep, no surprise HTML.
MCP server: Run npx @javinfo/mcp and your agent calls /movie and /query as native MCP tools — no glue code.

/llms.txtRFC 8288 Link headersnpm: @javinfo/mcp

// 05 — pricing

Simple access, free to start

Full data on every request — you pay per lookup, not per tier. Top up in USDC and manage keys in the app.

Free

1,000+ free lookups to start
No credit card required
60 requests / min
All providers, full fields

Pay as you go

Popular

$0.20 / 1,000 queries

$0.50 per 1,000 /movie · $0.20 per 1,000 /query
Top up in USDC, credited 1:1
$50+ top-ups earn a 5–50% bonus
Same full data, no gating

We accept crypto only.

// 06 — faq

Frequently asked

What javinfo does, how access works, and what happens to request data.

See all questions

What is javinfo?

javinfo is a JAV metadata search API. It checks registered providers in priority order and returns the first valid match in one response shape, no matter which provider answered. It indexes metadata only; it does not host or stream video.

How do I get access?

Where does the metadata come from?

javinfo uses a unified metadata database plus scrapers for other providers. You send one query; javinfo handles the source lookup and returns the result in the same format each time.

Does javinfo host or stream videos?

No. javinfo returns metadata: titles, cast, release info, and cover art references from public third-party providers. There are no streams and no hosted files here. It is a search tool, not a tube site.

What does it cost?

Pricing is usage-based and managed in the app: $0.0005 per result on /movie and $0.0002 on /query. Top up in USDC credited 1:1, and top-ups of $50+ get a random 5–50% bonus. It's also available on RapidAPI.

// 07 — blog

Latest from the blog

Guides and field notes on JAV, DMM and provider APIs.

View all posts

Ship the metadata layer today

Every source, one schema, two endpoints. Grab a free key and make your first call in minutes.