For AI agents and developers

Everything on Ghungroo can be read by software. It is public and read-only, and every answer carries the address of the page it came from so it can be cited.

MCP server

Point any MCP client at https://ghungroo.cleartrust.cc/mcp (Streamable HTTP). No key is needed. Tools: search, get_dance, list_dances, random_dance, dance_of_the_day, dancing_today, get_dance_day, list_dance_days, list_countries, get_country_dance, list_cities, list_ways_of_dancing, get_way_of_dancing, define_term, latest_news, stats, site_info.

{ "mcpServers": { "ghungroo": { "type": "http", "url": "https://ghungroo.cleartrust.cc/mcp" } } }

JSON API

The same tools answer plain HTTP: GET https://ghungroo.cleartrust.cc/api/v1/<tool>?name=value. For example /api/v1/get_dance?slug=bharatanatyam, /api/v1/list_dances?country=IN&form=folk&limit=20 or /api/v1/dancing_today. The list of tools with their parameters is at /api/v1/tools.

For publication networks

/api/v1/site_info returns a card for this publication: name, address, topics, counts, feeds, ad slots and contact. ClearTrust's network tools (the CPN portal, Footfall, Yieldo and Assay) read it.

Limits

90 calls a minute without a key. For more, ask for a key at ghungroo@cleartrust.cc and send it as Authorization: Bearer <key>.

For crawlers

llms.txt, llms-full.txt, sitemap.xml and an RSS feed of the dance of the day and new pages. AI crawlers are welcome in robots.txt.

Using the content

Quote freely with a link to the page. Dance pages are derived from Wikipedia and are available under CC BY-SA 4.0.