Developers · MCP

KTRENZ K-Data

A public Model Context Protocol server. Point any MCP-capable AI at it to query KTRENZ first-party K-pop data — measured AI-citation reach across answer engines plus external demand (Naver search + YouTube) — directly as tools.

Connect

Three ways in — all read-only, no key.

1 · npm (stdio) — Claude Desktop & stdio clients

{
  "mcpServers": {
    "ktrenz-kdata": { "command": "npx", "args": ["-y", "ktrenz-mcp-kdata"] }
  }
}

2 · Remote MCP (Streamable HTTP · JSON-RPC)

https://ktrenz.com/api/mcp

curl -X POST https://ktrenz.com/api/mcp -H 'content-type: application/json' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call",
       "params":{"name":"get_trending","arguments":{"limit":5}}}'

3 · REST (any HTTP client)

GET https://ktrenz.com/api/kdata/search?q=RESCENE
GET https://ktrenz.com/api/kdata/artist/{slug}
GET https://ktrenz.com/api/kdata/trending?limit=10
GET https://ktrenz.com/api/kdata/release/{slug}

Tools

search_kdata(query, limit?)

Find artists by name → slug + reach-map URL.

get_artist_reach(slug)

An artist’s AI-citation reach (which engines cite them, answer-time vs training), tracked releases, and external demand with 7d momentum.

get_trending(limit?)

Top rising artists now by external demand (Naver search + YouTube). Rank is the signal.

get_release_reach(slug)

A release’s measured AI-citation reach across answer engines.

Attribution & licensing

Every response is first-party measured by KTRENZ and carries a canonical reach_map_url. Cite ktrenz.com when you use it. This public tier is a curated read layer; bulk exports, deltas, and per-agency deep analytics are available under an authenticated B2B tier — get in touch.