Price, market cap, volume and all-time extremes for the ranked coin universe, with 24h / 7d / 30d change already computed.
price and rank are on every record; market_cap on 98% and volume on 99%. The percentage-change fields are precomputed across three windows, so a mover list needs no history calls.
Every field the corpus returns for one record. Real shape — walked corpus-wide, not sampled.
A curl that runs against the live API right now, plus the MCP tool names for the same data.
MCP tools
get_market_coinsget_market_coins_detailget_market_coins_historyget_market_trendingget_market_trending_detailThree concrete questions this capability answers that a general crypto API does not.
price_percent_change_24h, _7d and _30d are already on the row, so ranking by momentum is a sort rather than a history walk.ath and atl sit beside price on 86% of rows, so "how far off the high" is one subtraction and needs no separate endpoint.slug is the join key that /kasandra/patterns/{coin}/ and the tag filters expect — bitcoin, never BTC.Required, not optional. Coverage gaps, staleness and parameter traps that were observed by running the curl above against the live server on 2026-09-18.
updated_at is per coin, and the spread is months.
Walked all 1,321 rows on 20 Sep 2026: updated_at spans 2026-05-25 to 2026-09-20 across 69 distinct days. The top of the book is fresh; the long tail is not. Read updated_at before treating a row as current, and do not present the set as a single market snapshot.
ath and atl are missing on 14% of rows.
1,138 of 1,321 carry ath, 1,134 carry atl — newer listings mostly. A drawdown view has to tolerate nulls rather than assuming the pair is always present.