VNDB API Skill

VNDB API Skill

This skill helps interact with the VNDB (Visual Novel Database) API v2 (Kana) to query information about visual novels, characters, releases, producers, staff, tags, traits, and user lists.

API Endpoint

Base URL: https://api.vndb.org/kana

Sandbox URL: https://beta.vndb.org/api/kana (for testing)

Authentication

Most endpoints work without authentication, but user list management requires a token:

1. Obtain a token from VNDB profile → Applications tab (or https://vndb.org/u/tokens)
2. Token format: xxxx-xxxxx-xxxxx-xxxx-xxxxx-xxxxx-xxxx
3. Include in requests: Authorization: Token <your-token>

Rate Limits

• 200 requests per 5 minutes
• 1 second of execution time per minute
• Requests > 3 seconds are aborted

Common Data Types

• vndbid: Identifier with prefix (e.g., v17 for visual novels, c123 for characters, r456 for releases)
• release date: "YYYY-MM-DD", "YYYY-MM", "YYYY", "TBA", "unknown", "today"

Available Endpoints

Simple Requests (GET)

Database Querying (POST)

All POST endpoints accept a JSON query object:

{
  "filters": [],
  "fields": "",
  "sort": "id",
  "reverse": false,
  "results": 10,
  "page": 1
}

User List Management

Requires authentication with listread or listwrite permission.

Filter Syntax

Filters use a 3-element array: ["field", "operator", "value"]

Operators: =, !=, >=, >, <=, <

Combine filters with and/or:

["and", ["lang", "=", "en"], ["rating", ">", 80]]

Common Filter Flags

• o: Supports ordering operators (>, <, etc.)
• n: Accepts null as value
• m: Single entry can match multiple values
• i: Inverting filter is not always equivalent to inverting selection

Python Script Usage (Recommended)

For a more robust and maintainable solution, use the provided Python script:

Location

scripts/​vndb_query.py

Requirements

• Python 3.7+ (uses only standard library, no external dependencies)

Quick Commands

# Search characters
python scripts/​vndb_query.py character "美雪"

# Search visual novels
python scripts/​vndb_query.py vn "Steins;Gate"

# Get VN by ID
python scripts/​vndb_query.py vn_id "v17"

# Get latest released games
python scripts/​vndb_query.py latest 5

# Get database stats
python scripts/​vndb_query.py stats

# Query user
python scripts/​vndb_query.py user "yorhel"

Advanced Usage

# Custom fields
python scripts/​vndb_query.py character "美雪" "name,original,description,vns.title" 10

# Generic endpoint query
python scripts/​vndb_query.py query vn '{"search": "悬疑"}' "title,rating" "votecount" 20

Available Commands

Field Reference

Character fields: name,original,image.url,image.thumbnail,description,vns.title,vns.id,traits.name

VN fields: title,alttitle,image.url,image.thumbnail,rating,votecount,released,developers.name,description,platforms,languages

Release fields: id,title,alttitle,released,platforms,languages,producers.name,producers.developer,minage

Producer fields: id,name,original,lang,type,description

Staff fields: id,name,original,lang,gender,aliases.name

Tag fields: id,name,aliases,category,description,vn_count

Trait fields: id,name,aliases,group_name,description,char_count

See references/​api_docs.md for complete field documentation.

curl Example Requests (Alternative)

Search Visual Novels

curl https://api.vndb.org/kana/vn \
  --header 'Content-Type: application/​json' \
  --data '{
    "filters": ["search", "=", "Steins;Gate"],
    "fields": "title, image.url, rating, description",
    "results": 5
  }'

Get VN by ID

curl https://api.vndb.org/kana/vn \
  --header 'Content-Type: application/​json' \
  --data '{
    "filters": ["id", "=", "v17"],
    "fields": "title, image.url, rating, description, tags.name"
  }'

Get Characters for a VN

curl https://api.vndb.org/kana/character \
  --header 'Content-Type: application/​json' \
  --data '{
    "filters": ["vn", "=", ["id", "=", "v17"]],
    "fields": "name, image.url, description"
  }'

Get User's List (Authenticated)

curl https://api.vndb.org/kana/ulist \
  --header 'Content-Type: application/​json' \
  --header 'Authorization: Token <token>' \
  --data '{
    "user": "u2",
    "fields": "id, vote, vn.title",
    "sort": "vote",
    "reverse": true,
    "results": 10
  }'

Response Format

{
  "results": [...],
  "more": false,
  "count": 100
}

• results: Array of matching entries
• more: True if more results available
• count: Total count (only if "count": true in request)

Error Codes

Tips

• Select only needed fields to avoid "Too much data selected" errors
• Use id filtering for efficient pagination instead of page parameter
• Batch ID lookups: ["or", ["id", "=", "v1"], ["id", "=", "v2"], ...]
• Cache schema data - it doesn't change often

Reference Documentation

For complete API documentation, see references/​api_docs.md in this skill folder.