# CryptoDataAPI — Full Endpoint Reference

> Complete inventory of all API endpoints for LLM consumption. 12 aggregated data sources, 100+ endpoints.

## Authentication

All endpoints (except `/api/v1/health` and `/api/v1/payments/plans`) require:

```
X-API-Key: cdk_live_yourkey
```

Base URL: `https://cryptodataapi.com`

## Rate Limits

- Free tier: 5 requests/min, 50 requests/day
- Pro tier: 30 requests/min, 10,000 requests/day
- Pro Plus tier: 60 requests/min, unlimited daily

---

## Historical & Time-Series Data

15+ endpoints return time-series data across four tiers: DB-persisted (730 days), remote-sourced (30-90 days), live on-demand (per request), and Market Intelligence (Pro plan, CoinGlass).

### DB-Persisted
- `GET /api/v1/market-health/history` — Health score + 11 components (daily, up to 730 days). Params: `days` (default 365)

### Remote-Sourced (30-90 day windows)
- `GET /api/v1/sentiment/stablecoins/remote-history` — Stablecoin mcap + daily inflows. Params: `days` (default 90, max 365)
- `GET /api/v1/derivatives/binance/history` — Funding, OI, L/S ratio. Params: `days` (default 30, max 90)
- `GET /api/v1/market-data/volume-history` — Total volume + buy ratio. Params: `days` (default 30, max 90)

### Live On-Demand
- `GET /api/v1/market-data/klines` — OHLCV candles (any Binance pair). Params: `symbol`, `interval` (1m-1M), `limit` (max 1000)
- `GET /api/v1/market-data/btc-price-history` — BTC price + 200D MA. Params: `days` (default 365, max 730)
- `GET /api/v1/derivatives/binance/funding-rates` — Binance perp funding history. Params: `symbol`, `limit` (max 100)
- `GET /api/v1/hyperliquid/funding-rates` — HL funding history. Params: `coin`, `limit` (default 30)
- `GET /api/v1/hyperliquid/candles` — HL OHLCV candles. Params: `coin`, `interval`, `limit`

### Market Intelligence (Pro plan)
- `GET /api/v1/market-intelligence/btc/cycle-indicators` — 11 cycle indicators (MVRV, NUPL, Puell, etc.). Params: `days`
- `GET /api/v1/market-intelligence/etf/{asset}/flows` — ETF inflows/outflows (btc, eth, sol, xrp)
- `GET /api/v1/market-intelligence/fear-greed-history` — Fear & Greed history (~2018+)
- `GET /api/v1/market-intelligence/stablecoin-history` — Stablecoin mcap timeseries
- `GET /api/v1/market-intelligence/grayscale/premium` — GBTC premium (historical, ends ~Jan 2024)
- `GET /api/v1/market-intelligence/coinbase-premium` — Coinbase BTC premium index (4h)
- `GET /api/v1/market-intelligence/borrow-interest` — Margin borrow rates (4h)

---

## Daily Snapshot

### GET /api/v1/daily
Full market overview in one call. Returns market health, fear & greed, derivatives, sentiment, macro, ETF flows, cycle indicators, and coin profiles.
- Parameters: `exchange` (optional, default: "hyperliquid") — filter coins: "hyperliquid", "binance_spot", "asterdex", or "all"
- Parameters: `format` (optional) — set to "markdown" for LLM-friendly plain text
- Parameters: `technical_detail` (optional, default: false) — include the full per-symbol `technical_regime.by_symbol` map (keyed by ticker; each entry has ma/bollinger/range_state/rsi). Omitted by default to keep the payload compact; always available point-in-time via `/backtesting/daily-snapshots/{date}`.
- Parameters: `signum_detail` (optional, default: false) — include the full per-symbol `signum_rgg.by_symbol` map (keyed by ticker; RED/GREY/GREEN `color` + signed trend `score` + ADX/DMI + `days_in_color` + `hl_symbol` join key). Omitted by default; always present point-in-time via `/backtesting/daily-snapshots/{date}`.
- Response: `{ market_health: {...}, fear_greed: {...}, derivatives: {...}, stablecoins: {...}, macro: {...}, etf_flows: {...}, cycle_indicators: {...}, technical_regime: {counts + top_in_squeeze/top_overbought/top_oversold/top_near_high/top_near_low + by_symbol}, signum_rgg: {green/grey/red counts + shares + net_breadth_pct + mean_trend_score + sentiment + top_green/top_red + by_symbol}, coins: [...], snapshot_time: "..." }`
- Tip: This is the single most efficient endpoint for AI agents. Start here.

### GET /api/v1/daily/prices
All Binance spot price pairs (~2,500 symbols).
- No parameters

### GET /api/v1/daily/hyperliquid
Hyperliquid perpetual data: prices, funding rates, and open interest (~229 assets).
- No parameters

### GET /api/v1/daily/hl-traders
Daily Hyperliquid trader snapshot — the scored top-trader leaderboard, tracked wallet positions, and tracking metadata. 503 until the first leaderboard refresh completes.
- No parameters

---

## Market Health

### GET /api/v1/market-health
Full dual-score market health with all 11 components.
- Parameters: `format` (optional) — "markdown" for plain text
- Response: `{ total_score, long_term_score, short_term_score, state, sentiment, long_term_sentiment, short_term_sentiment, components: [...], long_term_components: [...], short_term_components: [...], details, indicators, timestamp }`
- States: bear_market, early_recovery, early_bull, confirmed_bull, topping_out

### GET /api/v1/market-health/summary
Lightweight scores + sentiment only.
- Parameters: `format` (optional) — "markdown" for plain text
- Response: `{ total_score, long_term_score, short_term_score, state, sentiment, long_term_sentiment, short_term_sentiment, timestamp }`

### GET /api/v1/market-health/components
All 11 component details.
- Response: `{ components: [{ name, score, weight, description, details }] }`

### GET /api/v1/market-health/component/{name}
Single component by name.
- Path parameter: `name` — component name string

### GET /api/v1/market-health/history
Health score history from database.
- Parameters: `days` (optional, default: 365, max: 730)
- Response: `{ history: [{ date, score, long_term_score, short_term_score, components }], count }`

### GET /api/v1/market-health/altcoin-breadth
Percentage of top altcoins above their moving average.
- Parameters: `ma_period` (optional, default: 200, range: 5-365)
- Response: `{ coins: [...], breadth_pct, ma_period }`

### POST /api/v1/market-health/refresh
Force recalculate health score (Pro tier only).

---

## Indicators (Pro tier required)

Computed trend and market-structure indicators. All endpoints in this section require a Pro (or Pro Plus) tier API key.

### GET /api/v1/indicators/signum-rgg
SIGNUM_RGG trend radar. Classifies every asset in the tradeable universe (all Hyperliquid perps + top 100 Binance USDT spot pairs by 24h quote volume, deduped) as RED (downtrend), GREY (sideways), or GREEN (uptrend) on the daily timeframe using ADX(14) with DMI directionality and a hysteresis band (20 <= ADX < 25 holds the prior color).
- Parameters:
  - `color` (optional): filter by `red`, `grey`, or `green`
  - `source` (optional): filter by `binance_spot` or `hyperliquid_perp`
  - `min_days` / `max_days` (optional): filter by days-in-current-color
  - `sort` (optional, default `days_in_color`): one of `days_in_color`, `pct_change`, `adx`, `symbol`
  - `order` (optional, default `desc`): `asc` or `desc`
  - `limit` (optional, default 250, max 500)
- Response: `{ items: [{ symbol, source, color, days_in_color, flipped_at, pct_change_since_flip, range: { low, high, width_pct } | null, adx, plus_di, minus_di, price }], count, universe_size, timestamp }`
- `range` is only populated for GREY assets (last 20-day high/low, width_pct = (high-low)/low*100).

### GET /api/v1/indicators/signum-rgg/{symbol}
Per-asset SIGNUM_RGG detail plus the last 60 days of color history for charting a trend-radar strip.
- Path parameter: `symbol` — bare ticker (e.g. `BTC`, `ETH`, `SOL`). `USDT` suffix is stripped automatically.
- Response: same fields as list item, plus `history: [{ date, color }]`, `rule_params: { period, upper, lower, range_lookback }`, and `timestamp`.

### POST /api/v1/indicators/signum-rgg/refresh
Force recompute the SIGNUM_RGG cache across the full universe. Normally refreshes once per day at 00:10 UTC.
- Response: `{ refreshed: true, count, timestamp }`

### GET /api/v1/indicators/technical
Technical / Structural regime overlay (Regime #14). Per-asset price-structure state across the SIGNUM_RGG universe (top Binance USDT pairs ∪ Hyperliquid perps, deduped). Acts as a universal overlay that pairs with every other regime — filter by indicator state to surface candidates.
- Computes four indicator families on daily klines (plus 1h for RSI):
  - **Moving averages**: SMA-50/100/200 + above/below flags + dist_from_X_pct. Tracks the most recent SMA-50 vs SMA-200 cross (`last_cross_200` = `golden` | `death`, with `days_since_cross_200`). Same for SMA-50 vs SMA-100.
  - **Bollinger bands** (20, 2σ): middle/upper/lower + `bandwidth_pct`. `in_squeeze` = current bandwidth ranks in the bottom 20th percentile of the trailing 90 days. `days_in_squeeze` = consecutive days the squeeze condition has held. `expanding` = bandwidth higher than 3 bars ago.
  - **Range state** (20d): low/high/width_pct + `position_pct` (0=at low, 100=at high) + `zone` (`near_low` | `mid` | `near_high`; near-edge thresholds at 15%).
  - **RSI** (14): daily + 1h values, state (`overbought` ≥70 / `oversold` ≤30 / `neutral`), `extreme_1d` / `extreme_1h` flags (≥80 or ≤20), `extreme_duration_bars` (consecutive recent 1d bars in the extreme zone).
- Parameters:
  - `source` (optional): filter by `binance_spot` or `hyperliquid_perp`
  - `ma_state` (optional): `above_200` | `below_200` | `breakdown_200` (recent SMA-50/200 death cross in last 10d) | `reclaim_200` (recent golden cross in last 10d)
  - `bb` (optional): `in_squeeze` | `expanding`
  - `range_zone` (optional): `near_low` | `mid` | `near_high`
  - `rsi` (optional): `overbought` | `oversold` | `extreme` (any side of 80/20)
  - `min_days_in_state` (optional): when combined with `bb=in_squeeze` or `rsi=...`, requires that many consecutive bars in state
  - `sort` (optional, default `symbol`): one of `symbol`, `price`, `rsi_1d`, `bb_bandwidth`, `squeeze_days`, `dist_from_200`, `position_in_range`
  - `order` (optional, default `desc`): `asc` or `desc`
  - `limit` (optional, default 250, max 500)
- Response: `{ items: [{ symbol, source, price, ma: {...}, bollinger: {...}, range_state: {...}, rsi: {...} }], count, universe_size, timestamp }`
- Cached for 30 minutes; refreshed every 30 min by the scheduler.

### GET /api/v1/indicators/technical/{symbol}
Per-asset Technical / Structural detail plus a compact 60d daily history series for sparkline rendering.
- Path parameter: `symbol` — bare ticker (e.g. `BTC`, `ETH`). `USDT` suffix is stripped automatically.
- Response: same fields as list item, plus `history: [{ date, close, rsi_14, bb_bw_pct, sma_200, above_200 }]` and `timestamp`.

### POST /api/v1/indicators/technical/refresh
Force recompute the Technical / Structural cache across the full universe.
- Response: `{ refreshed: true, count, timestamp }`

---

## Coins

### GET /api/v1/coins
List coins paginated by market cap rank.
- Parameters: `page` (default: 1), `limit` (default: 250)

### GET /api/v1/coins/search
Search coins by name or symbol.
- Parameters: `q` (required) — search query

### GET /api/v1/coins/top
Top N coins by market cap.
- Parameters: `n` (default: 10)

### GET /api/v1/coins/categories
All unique coin categories (raw CoinGecko tag names).

### GET /api/v1/coins/category-groups
Curated coin-category themes — 20+ major categories (AI, Layer-1/Layer-2, DeFi, DEX, lending, liquid staking, stablecoins, exchange tokens, meme, privacy, RWA, NFT, gaming, metaverse, DePIN, oracles, interoperability, storage, Bitcoin/Solana/Cosmos ecosystems, PoW, derivatives). Each group has a plain-English description and its resolved coin list (live price, market cap, rank, 24h/7d change), sorted by market-cap rank. Powers the /coin-categories page. Unlike /coins/categories (raw tag names), this is an editorial set.
- Parameters: `limit` (max coins per group, default 25)
- Response: `{ categories: [{ id, name, description, coin_count, coins: [{ symbol, name, current_price, market_cap, market_cap_rank, total_volume, price_change_24h_pct, price_change_7d_pct, image_url }] }], count }`

### GET /api/v1/coins/{symbol}
Detailed coin profile by symbol (uppercase).
- Path parameter: `symbol` — e.g., BTC, ETH, SOL
- Response: `{ symbol, name, current_price, market_cap, market_cap_rank, price_change_24h_pct, price_change_7d_pct, price_change_30d_pct, total_volume, circulating_supply, max_supply, ath, ath_date, atl, atl_date, categories, description }`

---

## Sentiment & Macro

### GET /api/v1/sentiment/fear-greed
Fear & Greed Index (multi-source averaged, 0-100).
- Parameters: `format` (optional) — "markdown" for plain text
- Response: `{ value, classification, sources, individual_values, timestamp }`
- Classifications: Extreme Fear (0-25), Fear (26-40), Neutral (41-60), Greed (61-75), Extreme Greed (76-100)

### GET /api/v1/sentiment/macro
Macro indicators: EUR/USD, gold, treasury yields.
- Response: `{ eur_usd, gold, treasury_2y, treasury_10y, treasury_30y, dxy, timestamp }`

### GET /api/v1/sentiment/stablecoins
Stablecoin market cap + 14d/90d flows.
- Response: `{ total_market_cap, market_cap_change_14d, market_cap_change_90d, stablecoins: [...] }`

### GET /api/v1/sentiment/stablecoins/history
Raw stablecoin market cap history timeseries.

### GET /api/v1/sentiment/stablecoins/remote-history
Daily stablecoin mcap + inflows from remote source.
- Parameters: `days` (default: 90, max: 365)

---

## NFT Trade Volume

All endpoints in this section require a Pro (or Pro Plus) tier API key; `/nfts/correlations` is Pro Plus.

The NFT track combines two sources:
- **DefiLlama** (`nft.llama.fi`): undocumented public endpoints (`/volume`, `/exchangeStats`). Gives current-snapshot 1d / 7d collection volumes and marketplace breakdown. Refreshed daily; the API server captures each snapshot into the local backtesting DB, so forward history builds one day at a time.
- **Dune Analytics** (`nft.trades`): optional weekly backfill, requires `DUNE_API_KEY`. Provides deep historical daily volume by chain × collection × category back to 2017 across Ethereum, Solana, and 10 EVM L2s. First run does a full backfill; subsequent runs only re-query a 10-day overlap window.

Collections are grouped by **chain** (Ethereum, Solana, Bitcoin Ordinals, Polygon, Base, Immutable, Ronin, Flow), **category** (pfp / art / gaming / collectible / music / domain / metaverse / ordinal / other), and **tier** — `og_blue_chip` / `mid` / `long_tail`. The seed taxonomy lives at `src/data/nft_collections.json` and is editorial.

### GET /api/v1/nfts/overview
Headline NFT trade-volume stats — daily total series, breakdowns by chain / tier / category, top live collections, marketplace share.
- Parameters:
  - `days` (optional, default 90, max 3650) — days of history to return
- Response: `{ total_series, by_chain, by_tier, by_category, totals: { volume_usd_24h, volume_usd_7d, volume_usd_30d, volume_usd_90d }, top_collections, marketplaces, refreshed_at, history_days }`

### GET /api/v1/nfts/volume
Daily NFT volume bucketed by chain, tier, category, or collection. Pro tier.
- Parameters:
  - `by` ∈ chain | tier | category | collection (default `chain`)
  - `from`, `to` — ISO dates (YYYY-MM-DD). Defaults to last 365d.
  - `granularity` (only `daily` supported today)
- Response: `{ by, granularity, points: [{ date, bucket, volume_usd }], start_date, end_date }`

### GET /api/v1/nfts/collections
Seeded NFT collection taxonomy.
- Response: `{ collections: [{ slug, name, chain, contract, category, tier, launched }] }`

### GET /api/v1/nfts/collections/{slug}
Per-collection daily volume time-series.
- Parameters: `days` (default 365, max 3650).
- Response: `{ meta: NFTCollection, series: [{ date, volume_usd, trades }] }`

### GET /api/v1/nfts/chains
Chain breakdown — chain name plus count of seeded collections on it.

### GET /api/v1/nfts/categories
Fixed category taxonomy.

### GET /api/v1/nfts/correlations
Pearson correlation of daily NFT total volume vs daily log-returns of named assets. Pro Plus.
- Parameters: `vs` (comma-separated, default `btc,eth,sol`), `window_days` (default 90).
- Response: `{ items: [{ asset, correlation, window_days, overlapping_days }], window_days, method, notes }`
- Notes: Pearson on log-returns, not a causal claim. Useful for sanity-checking whether NFT activity is tracking the broader crypto cycle in a given window.

---

## Derivatives (Binance Futures)

### GET /api/v1/derivatives/binance/funding-rates
Binance perpetual funding rate history.
- Parameters: `symbol` (default: "BTCUSDT"), `limit` (default: 30, max: 100)

### GET /api/v1/derivatives/binance/open-interest
Binance open interest + 30-day trend.
- Parameters: `symbol` (default: "BTCUSDT")

### GET /api/v1/derivatives/binance/long-short-ratio
Binance long/short account ratio.
- Parameters: `symbol` (default: "BTCUSDT")

### GET /api/v1/derivatives/binance/summary
All-in-one Binance derivatives summary for a symbol.
- Parameters: `symbol` (default: "BTCUSDT")

### GET /api/v1/derivatives/binance/history
Daily derivatives history (funding, OI, L/S).
- Parameters: `days` (default: 30, max: 90)

### GET /api/v1/derivatives/funding-rates
Cross-exchange funding rates (Binance + Hyperliquid + aggregators).
- Parameters: `coin` (default: "BTC")

### GET /api/v1/derivatives/open-interest
Cross-exchange open interest.
- Parameters: `coin` (default: "BTC")

### GET /api/v1/derivatives/summary
Combined cross-exchange derivatives overview.
- Parameters: `coin` (default: "BTC")
- Parameters: `format` (optional) — "markdown" for plain text

---

## Market Intelligence

### GET /api/v1/market-intelligence/btc/cycle-indicators
All 8 BTC cycle indicators with scores and zone classifications.
- Indicators: MVRV Z-Score, NUPL, Puell Multiple, Pi Cycle Top, 200-Week MA Multiple, Reserve Risk, RHODL Ratio, Thermocap Multiple

### GET /api/v1/market-intelligence/btc/cycle-indicators/{indicator}
Single BTC cycle indicator by name.

### GET /api/v1/market-intelligence/etf/{asset}/flows
ETF flow history for BTC, ETH, SOL, or XRP.
- Path parameter: `asset` — btc, eth, sol, or xrp

### GET /api/v1/market-intelligence/etf/btc/aum
BTC ETF total assets under management.

### GET /api/v1/market-intelligence/liquidations
Cross-exchange liquidation data (long + short volumes 1h/4h/12h/24h).
- Parameters: `exchange` (optional, default: "hyperliquid") — coin-set filter: "hyperliquid", "binance_spot", "asterdex", or "all"
- Parameters: `symbol` (optional) — filter to a single coin
- Parameters: `type` (optional), `limit` (default: 250)
- With `exchange=asterdex`, response includes AsterDEX-native rows (tagged `source: "asterdex"`) for AsterDEX-only symbols (e.g. LAB, ASTEROID) alongside the CoinGlass cross-exchange aggregate.

### GET /api/v1/market-intelligence/liquidations/by-exchange
Liquidations broken down by exchange.

### GET /api/v1/market-intelligence/options
BTC options data: OI, volume, put/call ratio, max pain.

### GET /api/v1/market-intelligence/exchange-balance
Exchange BTC balance and flow data.

### GET /api/v1/market-intelligence/coinbase-premium
Coinbase premium index (BTC price premium on Coinbase vs other exchanges).

### GET /api/v1/market-intelligence/funding-rates
Cross-exchange funding rates from aggregator.
- Parameters: `exchange` (optional), `type` (optional), `limit` (default: 250)

### GET /api/v1/market-intelligence/open-interest
Cross-exchange open interest from aggregator.
- Parameters: `exchange` (optional), `type` (optional), `limit` (default: 250)

### GET /api/v1/market-intelligence/grayscale/holdings
Grayscale fund holdings across all products.

### GET /api/v1/market-intelligence/grayscale/premium
Grayscale BTC premium/discount history.

### GET /api/v1/market-intelligence/taker-buy-sell
Taker buy/sell volume ratio by exchange for multiple coins (BTC, ETH, SOL, DOGE, XRP, ADA).
Optional query param: `?symbol=BTC` to filter to a single coin.

### GET /api/v1/market-intelligence/borrow-interest
Margin borrow interest rate history.

### GET /api/v1/market-intelligence/fear-greed-history
Fear & Greed index history with dated entries.

### GET /api/v1/market-intelligence/stablecoin-history
Stablecoin market cap history timeseries.

### GET /api/v1/market-intelligence/status
Collector status and rate limit usage.

---

## On-Chain Intelligence

Leading on-chain signals sourced from QuickNode RPC nodes (ETH/Tron/BSC/Base/BTC/Sol), mempool.space, Ethplorer, and CoinMetrics Community API. All data is read directly from chain nodes, not scraped from third-party UIs.

### GET /api/v1/on-chain/stablecoin-reserves
CEX stablecoin reserves across ETH/Tron/BSC. Aggregate USDT/USDC balances in known hot/cold wallets of Binance, Coinbase, Kraken, Bybit, and OKX. Returns by_exchange / by_chain / by_symbol breakdown + total_usd.

### GET /api/v1/on-chain/stablecoin-reserves/dry-powder
Z-scored dry-powder signal. Returns "accumulating" (z > +1, bullish), "neutral", or "depleting" (z < -1, bearish) — based on a rolling 30-day baseline.

### GET /api/v1/on-chain/exchange-flows/{symbol}
Net Transfer flow to/from CEX wallets per symbol. The top-level `windows` are **summed across every chain the symbol exists on**; the response also includes a per-chain `by_chain` breakdown (same window shape, only chains with data) and a `chains_covered` list — matching the backtesting archive so live and historical snapshots have the same structure. Current coverage:

- **Ethereum** (50 tokens): USDT, USDC, DAI, WBTC, WETH, LINK, UNI, AAVE, MKR, LDO, ARB, MATIC, PEPE, SHIB, RNDR, INJ, FET, GRT, IMX, MANA, SAND, AXS, APE, BLUR, ENS, COMP, SNX, CRV, SUSHI, 1INCH, LRC, CHZ, GALA, PYUSD, FRAX, ONDO, WLD, ENA, USDE, PENDLE, W, TIA, STRK, ETHFI, EIGEN, RPL, NEAR, FXS, BNT, OCEAN.
- **BSC** (9 tokens): USDT, USDC, BUSD, CAKE, BTCB, ETHB, FDUSD, TWT, DAI.
- **Base** (7 tokens): USDC, WETH, CBBTC, AERO, DEGEN, VIRTUAL, BRETT.
- **Arbitrum** (5 tokens): USDC, USDT, WETH, ARB, GMX.
- **Optimism** (4 tokens): USDC, USDT, WETH, OP.
- **Tron** (2 tokens): USDT, USDC.
- **Solana** (13 symbols via Helius enhanced API, wallet-centric scan): SOL native, USDC, USDT, JUP, JTO, BONK, WIF, POPCAT, MEW, PYTH, RAY, ORCA, HNT.

Returns 1h/6h/24h/7d windows, per-exchange 24h breakdown (binance, coinbase, kraken, bybit, okx), `by_chain` + `chains_covered`, plus rolling spike_zscore_1h and spike_zscore_24h. Positive net = deposits (often bearish leading), negative net = withdrawals (often bullish). Optional ?exchange=binance filter. Refresh cadence 5min. 90d historical depth available via the backtesting archive (with the same per-chain breakdown under `exchange_flows.by_chain.<chain>`).

**Solana caveat (withdrawals-only bias):** Solana CEX deposits use per-user token accounts owned by the hot wallet; user-signed deposit transactions don't appear when scanning the hot wallet directly via Helius's address-transactions API. As a result, SPL/SOL inflow numbers are systematically low. Outflow (withdrawal) numbers are accurate. Net flow is biased toward "outflow" for SPL tokens on Solana. EVM chains (ETH/BSC/Tron) are not affected — those scan token contracts and capture both directions correctly.

**Native ETH caveat (live-only):** The `ETH` symbol (native, distinct from `WETH` ERC-20) is tracked via eth_getBalance delta polling at each CEX hot wallet. No historical backfill — flow data accumulates forward from collector start. Use `WETH` for backtest periods before collector deploy.

### GET /api/v1/on-chain/exchange-flows/spike-alerts
Recent large transfers (≥ min_amount, default $1M USD) to/from tracked CEX wallets across ETH and BSC. Each alert: chain, symbol, exchange, direction (inflow/outflow), amount, counterparty address. ?min_amount and ?limit (max 500) query params.

### GET /api/v1/on-chain/miners/reserves
BTC miner pool reserves + 24h/7d/30d net flow per pool. Tracks Foundry USA, AntPool, F2Pool, ViaBTC, Binance Pool reward addresses via mempool.space. Negative net = miners selling (bearish), positive net = accumulation. Refresh cadence 30min.

### GET /api/v1/on-chain/miners/hash-ribbon
Hash Ribbon indicator (Charles Edwards, 2019): 30-day MA vs 60-day MA of BTC hashrate. State machine: "capitulation" (30dMA < 60dMA), "recovery" (just crossed back above within 14d — historically the strongest BTC bottom signal), or "normal". Includes ma30_eh, ma60_eh, ratio, days_in_state. Refresh cadence 1h.

### GET /api/v1/on-chain/dormancy/btc
BTC dormancy & supply-shock signals via CoinMetrics Community API. Returns MVRV with zone classification (capitulation < 1.0, accumulation 1.0-1.5, neutral 1.5-2.5, elevated 2.5-3.5, euphoria > 3.5), plus mvrv_7d_change_pct, active_addresses, issuance_usd_24h, and chain-wide exchange_net_flow_usd_24h as an independent cross-check on the QuickNode-derived ERC-20 flow endpoints.

### GET /api/v1/on-chain/whales
🚧 **Coming soon (temporarily disabled).** Top-100 non-CEX holders snapshot across USDT, USDC, WBTC, WETH on Ethereum. CEX wallets filtered out via the address registry. Per-symbol: non_cex_holder_count, total_balance, total_share_pct, top_5 holders, 24h/7d/30d deltas, accumulation_signal.

### GET /api/v1/on-chain/whales/{symbol}
🚧 **Coming soon (temporarily disabled).** Per-token whale view for one of USDT / USDC / WBTC / WETH with full top-5 list and delta payload.

### GET /api/v1/on-chain/whales/accumulation-score
🚧 **Coming soon (temporarily disabled).** Aggregate whale accumulation verdict across all tracked ERC-20s. Returns directional signal (accumulating / neutral / distributing / unknown) plus per-category counts. Useful as a single boolean for an AI agent to pivot bias on.

### GET /api/v1/on-chain/whales/accumulation-score/{symbol}
🚧 **Coming soon (temporarily disabled).** Per-symbol whale accumulation signal with deltas.

### GET /api/v1/on-chain/score
Composite 0-100 on-chain health score with per-component breakdown. Weights: MVRV 25%, Hash Ribbon 20%, dry-powder 15%, whale accumulation 15%, exchange flow 15%, miner flow 10%. Sentiment bands: bullish (≥70), leaning_bullish (55-69), neutral (45-54), leaning_bearish (30-44), bearish (<30). Components with no data yet contribute neutral 50 so the score remains stable across cold deploys.

---

## Liquidity / Market Depth

Per-minute Hyperliquid L2 book snapshots across the top-25 perps (by 24h notional volume), joined to the cached OI snapshot. Powers the Liquidity / Market Depth regime on the /regimes page.

### GET /api/v1/liquidity/depth
Current per-coin depth/spread snapshot for the tracked universe. Each record: mid_price, spread_bps, bid/ask depth USD at {10, 25, 50, 100}bps from mid, total_depth_25bps_usd, imbalance_10bps (book skew within 10bps), and joined open_interest_usd. Refresh cadence: 60s.

### GET /api/v1/liquidity/oi-divergence
Per-coin OI vs price change across 1h/4h/24h windows, ranked by 4h divergence (oi_change_4h_pct − price_change_4h_pct). Positive divergence = OI rising faster than price = fragile positioning. Coins without enough history for a given window report null for that window.

### GET /api/v1/liquidity/regime
Per-coin liquidity regime label + market-wide aggregate + composite fragility score. Per-coin regimes (applied in priority order): post_cascade_impaired, depth_withdrawal, oi_price_divergence, deep_book, neutral. Each entry includes confidence (0-1), low_confidence flag (true when <60min of buffered samples), baseline_minutes, and the raw signals that triggered the rule. Coins below the confidence threshold still classify but flag low_confidence. Pro Plus only.

### GET /api/v1/liquidity/regime/score
Composite 0-100 market structure score derived from the regime mix (higher = healthier). Sentiment bands: healthy (≥70), leaning_healthy (55-69), neutral (45-54), leaning_fragile (30-44), fragile (<30). Score weights vs baseline 50: +30 × pct_deep_book, −25 × pct_divergent, −30 × pct_withdrawing, −20 × pct_impaired. Trimmed companion to /liquidity/regime — same composite, no per-coin list. Pro Plus only.

### GET /api/v1/liquidity/depth/{coin}
Per-coin rolling depth history (up to 24h of 1-min samples). Query param: minutes (default 60, max 1440). Returns the same per-record shape as /liquidity/depth, ordered oldest→newest. Pro / Pro Plus only.

### GET /api/v1/volatility/regime
Per-asset realized-volatility regime (Regime #13) across the SIGNUM_RGG universe (top Binance USDT pairs ∪ Hyperliquid perps, deduped). A risk-sizing overlay computed from daily klines — fully backfillable, no new external data source.
- Realized vol via three estimators, annualized for a 24/7 market (×√365): close-to-close (log returns), Parkinson (high-low range), Garman-Klass (full OHLC, the headline). Computed over 7d and 30d windows.
- `vol_pctile_30` / `vol_pctile_7`: where current 30d / acute 7d vol ranks within its own trailing 90d distribution. Low percentile = the calm before expansion.
- `rv_z_30` / `rv_z_7`: realized-vol z-score (how many standard deviations current 30d / 7d vol sits from its own trailing-90d mean) — the magnitude companion to the percentile rank, for a continuous de-rate rather than a boolean gate.
- `term_structure_ratio` = 7d / 30d vol (>1 = front rising / vol breaking out).
- `vol_target_multiplier` = target_vol(60%) / current_vol, capped to [0.25, 3.0] — the scalar a vol-targeting strategy multiplies base size by (size shrinks as vol rises).
- `days_compressed` / `days_shock`: consecutive days in the low-vol / acute-spike condition. `days_in_regime` is the run-length of the *current* regime across all five states, with `prev_regime` (the regime immediately before the run) and `regime_changed` (true if the latest bar flipped) — so a strategy can detect the cross into a new regime point-in-time without holding its own prior-day state.
- `is_stable` (top-level per-symbol flag): true for known stablecoins or heuristically pegged tokens (priced ≈$1 with peg-like ultra-low vol). Lets consumers avoid over-weighting a peg in a "size up the calm coins" loop.
- Per-asset regime (priority order, first match wins): `vol_shock` (7d vol ≥90th pctile), `expanding` (term ratio ≥1.15 and 30d pctile ≥50), `compressed` (30d vol ≤20th pctile — dry tinder), `mean_reverting` (term ratio ≤0.85 and 30d pctile ≥60), `normal` (fallthrough).
- `vol.rv_1h` is a live-only intraday field; backfilled historical days land null.
- Parameters: `source`, `regime`, `sort` (default vol_pctile_30; one of symbol, rv_gk_30, rv_gk_7, vol_pctile_30, term_structure_ratio, vol_target_multiplier, days_compressed), `order` (default desc), `limit` (default 250, max 500).
- Response: `{ items: [{ symbol, source, price, regime, is_stable, vol: {...} }], count, universe_size, timestamp }`. Cached 30 min; refreshed every 30 min by the scheduler.

### GET /api/v1/volatility/regime/score
Market-wide volatility-stress composite (0-100; higher = more stress → size down) derived from the regime mix. Sentiment bands: stressed (≥70), elevated (55-69), normal (45-54), calm (30-44), dormant (<30, the dangerously-quiet compressed-vol state). Score weights vs baseline 50: +35 × pct_shock, +20 × pct_expanding, −20 × pct_compressed, +5 × pct_mean_reverting. Includes median_vol_pctile_30 + regime aggregate. Also returns `target_vol` (the 60% numerator behind the per-asset multiplier) and `gross_exposure_multiplier` (0.5–1.0, the universe-wide analog of the per-asset vol-target multiplier: ~1.0 in a calm tape, ramping to ~0.5 at max vol stress — a single scalar a market-wide gross overlay multiplies into). Trimmed companion to /volatility/regime — no per-coin list.

### GET /api/v1/volatility/regime/{symbol}
Per-asset Volatility regime detail plus a compact 60d daily history series (close, 30d close-to-close vol, 30d Garman-Klass vol) for sparkline rendering. Path param: bare ticker (USDT suffix stripped). Pro / Pro Plus only.

### POST /api/v1/volatility/regime/refresh
Force recompute the Volatility regime cache. Pro / Pro Plus only.

### GET /api/v1/quant/market
Whole-market regime + probability object from a Hidden Markov Model trained on years of hourly bars. Regime is one of 6 states (strong_trend_bull, strong_trend_bear, range_low_vol, choppy_high_vol, vol_spike, squeeze) with calibrated confidence and candles_in_regime. Probability buckets at the requested horizon (`horizon=4h|24h`, default 24h): directional (strong_down/mild_down/flat/mild_up/strong_up), volatility (low/medium/high), liquidation_risk, funding (falls/stable/rises), breadth (contracting/neutral/expanding — market scope only), open_interest, and regime_transitions (stays_same + to_<label> probabilities from exact transition-matrix powers). The market-scope liquidation_risk bucket is depth-aware: driven by live order-book fragility (top-10 spread/depth/imbalance — thin book + wide spread = higher forced-liquidation odds) rather than just realized volatility, tagged in explain.heads_provenance (e.g. depth_fragility:thin); the per-coin liquidation_risk stays volatility-derived. Each bucket carries its own confidence (entropy-based, degraded by data staleness/coverage). The 24h horizon adds `tomorrow`: seeded Monte Carlo stats (N=1000 paths — cumulative-return quantiles p5..p95, p_visit_vol_spike, p_drawdown_gt_5pct, label occupancy; deterministic seed = reproducible). The `explain` block exposes every adjustment: per-feature z-scores, raw HMM state posteriors, the state→label map, `horizon_label_distribution` (the FULL 6-regime distribution projected to the requested horizon via exact transition-matrix powers — the untrimmed now→horizon grid behind regime_transitions), the hysteresis rule + pending challenger, raw vs calibrated confidence, head provenance, and directional thresholds resolved to concrete % at current vol. Refreshed every 15 minutes. Pro tier (market-wide; the per-coin objects at /quant/coins* are Pro Plus).

### GET /api/v1/quant/coins
Trimmed regime summary for every active Hyperliquid perp (~200+) — the heatmap feed. Each item: regime (label/confidence/candles_in_regime), p_direction_up and p_direction_down (the two bullish / two bearish buckets respectively; up + down + flat = 1, so never treat 1 − p_up as fall odds), top_transition {to, p}, oi_usd, and data-quality flags: insufficient_history (<60d of bars, confidence halved), status warming_up (<30d, no regime call yet), new_listing (<7d on HL). Parameters: `horizon`, `regime` (filter by label), `sort` (oi | confidence | symbol | p_up, default oi), `order`, `limit` (max 300). Full probability matrices live at /quant/coins/{symbol}. Pro Plus only.

### GET /api/v1/quant/coins/risk
Bulk per-coin risk model in one call — a batched join of /quant/coins/{symbol} and /volatility/regime/{symbol} across the whole quant universe (majors BTC/ETH/SOL included), built so consumers don't fan out ~2 calls per perp every cycle. Response: `{ as_of, horizon, items: [...], count, universe_size }`; each item: `symbol`, `regime` {label, confidence}, `vol_target_multiplier` (0.25–3.0), `vol_pctile_30`, `rv_24h` (volatility regime `rv_1h`, live intraday RV; null on backfilled/illiquid), `probabilities` with the volatility/liquidation_risk/funding/open_interest buckets (verbatim from /quant/coins/{symbol}; directional omitted — get it from /quant/coins), and `meta` {status, insufficient_history, new_listing}. Coins with insufficient history or new listings still appear (flagged in meta) so callers can mark them ineligible. Values match the per-symbol endpoints for the same symbol/horizon — a batched view, not new modelling. Parameter: `horizon` (4h | 24h, default 24h). Pro Plus only.

### GET /api/v1/quant/coins/{symbol}
Full per-coin probability object — same shape as /quant/market minus breadth — conditioned on the whole-market regime: the coin's transition matrix is the market-posterior-weighted mixture of per-market-label matrices (A_mix = Σ P_market(m)·A_m), so SOL can read squeeze while the market reads range, and `explain.transition_mixture_weights` shows exactly how much each market label contributed. Includes `recent_regimes` (last ~48 inference rows: timestamp, regime, confidence) for sparkline rendering. Coins under 60d of history return the object flagged insufficient_history rather than 404. Path param: HL coin name (BTC, SOL, kPEPE). Pro Plus only.

### GET /api/v1/quant/history
Point-in-time record of every probability the engine emitted — the verification surface. Each row: flattened probability columns, regime + confidence, label posteriors, `model_version` (which artifact produced it), and `backfilled` (false = live 15-minute inference recorded at the time; true = retro-computed at 1h cadence by the backfill script). Use it to audit our calls against realized outcomes. Parameters: `scope` ('market' or coin symbol), `start` (ISO or unix ms, required), `end`, `horizon`, `limit` (max 2000), `format` (json | csv). Pro Plus only.

### GET /api/v1/quant/regimes/history
Bulk download of the full market 6-regime history as a Parquet file (returns a pre-signed URL, 1h expiry). One hourly row per bar from 2020 to yesterday, market scope, each carrying the full probability distribution over all 6 regimes at three horizons — now, 4h and 24h (columns `p_now_0..5`, `p_4h_0..5`, `p_24h_0..5`, keyed by regime id; see `regime_map` in the response), plus `regime_id`/`regime_label`/`regime_conf`, `candles_in_regime`, `model_version` and a `source` column marking the deep Binance-archive segment vs the live-Hyperliquid tail. The deep history is labelled by the same model that serves live, run over the public Binance archive it trained on (no hindsight); `p_4h`/`p_24h` are exact transition-matrix projections (alpha^T A^n) of the current posterior, not realized outcomes. Rebuilt daily at 02:35 UTC. Use this for backtesting instead of paginating /quant/history. No parameters. Pro Plus only.

### GET /api/v1/quant/timeline
Daily market regime label from 2019 to now, generated at training time by running the final model over its full training history — the COVID crash, 2021 bull, LUNA/FTX bear and 2024 ETF bull all labeled by the same model that serves live, with no hindsight relabeling. Regenerated on every retrain. Parameter: `start` (ISO date filter). Pro Plus only.

### GET /api/v1/quant/positioning
Per-coin trader positioning by account type, built from the full Hyperliquid account universe (every account ≥ $100k that holds a position, refreshed ~5-min via the live liquidation-map collector). For each coin we aggregate long/short/net/gross notional into trader-type buckets — `market_maker`, `whale`, `other`, and `all` — plus the distinct account count per bucket and the current mark. Trader type comes from a behavioral classifier (monthly turnover ÷ account value, number of coins quoted, average leverage, consistent profitable timeframes + all-time PnL, and a curated known-MM/vault list). The `market_maker` bucket is the basis for the dealer-gamma view at /quant/gex. Parameter: `symbol` (filter to one coin). Pro Plus only.

### GET /api/v1/quant/gex
Dealer positioning / "Gamma Exposure" — the perpetuals analog of options GEX, built from market-maker accounts. Per coin: `mm_net_delta` (net MM inventory long/short = the dealer book), `gamma_profile` (the MM liquidation-density binned by price = where market-maker forced unwinds cluster, the short-gamma acceleration zones), and `gamma_flip` (the price where cumulative MM long-liq-below crosses short-liq-above — the inflection). `gamma_flip` is capped to within ±30% of mark and is `null` when no liquidation crossover falls in that band (previously it could report a stale level hundreds of % from mark — treat `null` as "no meaningful flip", not zero). Each coin also carries a composite **`regime`** flag — `amplify` (short-gamma analog: crowded book + dense clusters near mark → forced flow follows through, don't fade), `dampen` (balanced, funding near baseline, no overhead clusters → fade extremes) or `transitional` (on the flip line) — with a signed `score`, a `confidence` (0–1, scaling with the MM account count + gross book behind the read so thin-sample regimes can be down-weighted), `fuel_side` (long_squeeze / short_squeeze / balanced) and a full `inputs` block (funding_rate/funding_apy/funding_state, premium, oracle_price, open_interest, oi_change_pct, cluster_density_near_pct, market_skew, and the realized-liquidation cascade signal with its spike_ratio; realized_liq now covers 1000x-multiplier meme perps too). Delisted HL markets (zero OI/funding) are excluded from the universe. Top-level `funding_rate`, `oracle_price`, `premium`, `open_interest(_usd)` and `dist_to_flip_pct` are lifted out for convenience, and `meta.maintenance_margin_tiers` gives the per-pair MM/max-leverage ladder. The full per-coin map is archived at 5-min cadence for backtesting as the `gamma_exposure` snapshot type (byte-parity with this endpoint); the sibling `liquidation_map` snapshot carries the full-universe composite regime. Honest framing: perps have no literal options gamma; this is a behavioral analog from MM inventory + liquidation-driven forced flow, and the MM set is a heuristic classification — treat it as a positioning signal, not an exact dealer book. Parameter: `symbol` (filter to one coin; returns the same envelope narrowed to `coins:{SYMBOL:{...}}`). Pro tier is scoped to BTC; full-universe coverage requires Pro Plus.

### GET /api/v1/quant/whales
Hyperliquid whale activity — the entire ≥$100k account universe (the same ~5-min-fresh set behind /quant/positioning and /quant/gex) rolled up into one read. Response: `summary` { accounts_tracked, universe_size, min_account_value_usd, by_class {market_maker, whale, other}, coins_held, total_long_usd, total_short_usd, total_net_usd, total_gross_usd, long_pct, long_short_ratio, net_bias (long/short/neutral) }, `top_coins` [ per coin: coin, mark, segment, gross_usd, long_usd, short_usd, net_usd, directional_net_usd (net excluding market-maker liquidity = the conviction read), long_pct, net_bias, n_accounts, n_long, n_short, dominant_side ] sorted by gross notional, and `meta` { segment: "perp", spot_status, venue, by_tag (raw classifier tag counts), note }. Scope is Hyperliquid perpetuals (the whale margin book); spot-wallet balances are not yet collected and the view is crypto-only (tokenized equities excluded). The heavy roll-up is cached in-service (~120s). No parameters. Pro tier is scoped to BTC (top_coins); full-universe coverage requires Pro Plus.

### GET /api/v1/quant/whales/history
Historic whale-activity movement — a daily time-series of aggregate whale positioning. Response: `points` [ per UTC day: date, ts_ms, accounts, long_usd, short_usd, net_usd, gross_usd, top_coins [{coin, net_usd}], source ("seed" | "live"), estimated (bool) ], plus `count` and `meta` { segment, venue, seeded_count, live_count, first_date, last_date, note }. Full-universe collection is recent, so the earlier portion of the series is modeled (estimated=true) — a seeded baseline anchored to the first live reading — and is replaced by observed daily snapshots as they accumulate; the seeded_count/live_count split tells you exactly how much is observed vs modeled. Parameter: `days` (trailing window, 7–540, default 180). Pro Plus only.

### GET /api/v1/quant/model
Model card + runtime status, open to any valid key: model_version, family, feature lists per scope, raw HMM state counts, walk-forward validation fold metrics straight from the artifact, artifact sha256, human approval stamp, and live runtime health (universe coverage, 7d log-likelihood drift vs the training baseline, last inference time). `loaded: false` means the quant data endpoints return 503 (no trained model deployed yet).

### GET /api/v1/quant/regimes
The 6-regime taxonomy used by every quant endpoint: id, label, display name, description, suggested algo-basket stance (e.g. squeeze → breakout algos primed, sizing up; vol_spike → pause all), and the regime color used on the /quant page. Open to any valid key.

### GET /api/v1/regimes
The 10-state long-horizon market-cycle model — the public-facing lifecycle taxonomy shown on the /regimes page, plus a live classification of which regime the market is in right now. Returns `regimes` (the taxonomy: id 1-10, machine `regime` id, `anchor` page slug, name, description, typical_duration, directional_bias) and `current`. The 10 states in cycle order: structural_shock_crisis (an emergency override, not a normal stage), established_bear, capitulation, bottoming_accumulation, early_recovery, btc_led_bull, broadening_bull, broad_bull_alt_expansion, speculative_euphoria, distribution_post_peak.
- `current` carries the detected `regime` + `name` + `id`, a plain-English `rationale`, and `signals` — the inspectable inputs the classifier composed: long/short market-health scores + the derived cycle condition, BTC dominance, OI change, long/short split, plus the volatility / meme / security composite scores and the quant HMM overlay. Any feed that's cold simply drops out of `signals` (graceful degradation).
- Two states expose an internal sub-signal: speculative_euphoria adds a 0-100 `euphoria_score` (and `stage` = alt_expansion | blow_off) to separate an ordinary alt expansion from a terminal blow-off; distribution_post_peak adds a `stage` = early_distribution | active_deleveraging.
- This is the long-horizon (weeks-to-months) counterpart to /quant, which is a short-horizon (24h) HMM read. Cached ~5 min. Open to any valid key.

### GET /api/v1/regimes/current
The `current` block above on its own — the trimmed companion to /regimes when you don't need the taxonomy. Returns the current regime id/name, rationale, optional stage / euphoria_score, the classifier `signals` and an `as_of` timestamp. Open to any valid key.

### GET /api/v1/trading-strategy-baskets
A practical catalogue of 50 top-level trading-strategy meta-baskets across 6 thematic groups — the same list shown on the /trading-strategy-baskets page. Returns `groups` (A–F, each with id, name and its baskets) plus `group_count` and `basket_count`. Each basket has a stable `slug` (the API contract + page anchor), a display `name` and a one-line `description`. The six groups: A Cycle & Market Leadership, B Funding/Leverage & Liquidation, C Technical & Market Structure, D Macro & Institutional Flows, E Crypto-Specific Catalysts, F Narrative & Sector Rotation. These are the top-level taxonomy; more granular setups are modelled as sub-strategies, signals or directional variants beneath them. Pro tier.

### POST /api/v1/quant/refresh
Force an immediate inference cycle instead of waiting for the next 15-minute tick. Pro Plus only.

### GET /api/v1/meme/regime
Per-asset Meme / Speculative lifecycle regime (Regime #3) over a curated meme-perp universe — the speculative subset (DOGE, SHIB, PEPE, WIF, BONK, FLOKI, POPCAT, …) intersected with the live SIGNUM_RGG universe. k-prefixed Hyperliquid listings (kPEPE, kBONK, kSHIB, kFLOKI, …) are resolved back to their base ticker so they're no longer dropped. Aimed at systematic perps trading: the label implies a holding duration, leverage level, and funding-cost tolerance.
- Trailing returns `ret_1d` / `ret_7d` / `ret_30d` (%), `vol_spike` (latest bar volume ÷ its 30d median), `rv_gk_30` (30d Garman-Klass realized vol) + `vol_pctile_30` (rank within trailing 90d), `sma20_extension_pct` (price distance above/below SMA-20).
- `funding_rate` (live Hyperliquid hourly funding), `funding_extreme` (≥0.05%/hr), `oi_usd` (live open interest). These are live-only and land null on backfilled historical days.
- `days_in_regime` (run-length of the current regime), `prev_regime`, and `regime_changed` — lets a strategy catch the cross into `ignition` point-in-time. The latest bar uses live funding; historical bars are price/volume-only, so funding-gated states have `days_in_regime` = 1.
- `hl_symbol` (per coin, also in the live items and the daily `by_symbol` archive): the exact Hyperliquid perp join key (e.g. PEPE → "kPEPE"), so consumers join without a k/1000-strip heuristic. Null when the coin has no HL perp.
- Per-asset regime (priority order, first match wins): `euphoric` (30d vol ≥80th pctile with 7d ≥+25%, or extreme positive funding while rising — blow-off), `distribution` (extended ≥+40% over 30d but 7d stalled ≤+5% with funding still positive-elevated — topping; live-only), `ignition` (7d ≥+15% backed by a ≥2× volume spike — entry), `bleeding` (7d ≤−15% — unwind), `dormant` (fallthrough).
- Price/volume/vol features are backfillable from daily klines; funding-gated states (`distribution`, the funding path of `euphoric`) only surface on the live snapshot.
- Parameters: `source`, `regime`, `sort` (default ret_7d; one of symbol, ret_7d, ret_30d, vol_spike, vol_pctile_30, sma20_extension_pct, funding_rate, oi_usd), `order` (default desc), `limit` (default 250, max 500).
- Response: `{ items: [{ symbol, source, price, regime, hl_symbol, meme: {...} }], count, universe_size, timestamp }`. Cached 30 min; refreshed every 30 min by the scheduler.

### GET /api/v1/meme/regime/score
Market-wide meme-hype composite (0-100; higher = frothier / more speculative heat) derived from the regime mix. Sentiment bands: euphoric (≥70), heating (55-69), neutral (45-54), cooling (30-44), dormant (<30). Score weights vs baseline 50: +30 × pct_euphoric, +20 × pct_ignition, +5 × pct_distribution, −20 × pct_bleeding, −15 × pct_dormant. Also returns `breadth_heating_pct` (share euphoric+ignition), `mean_pairwise_correlation` (cross-meme co-movement of the last 14 daily returns), `meme_season` (broad heating + correlated pack move), `median_funding_rate` (live), and the regime aggregate. `fresh_issuance` is a live-only froth gauge over the long tail — DEX-Screener promotion breadth/spend joined with GoPlus rug density: `{ froth_score (0-100 heuristic), promoted_count, total_promotion_spend, rug_sampled, rug_high_risk_pct, avg_security_score }` (null if unavailable). It sits alongside the backfillable `composite_score`, not inside it. Trimmed companion to /meme/regime — no per-coin list.

### GET /api/v1/meme/regime/{symbol}
Per-asset Meme regime detail plus a compact 60d daily history series (close, 30d Garman-Klass vol) for sparkline rendering. Path param: bare ticker (USDT suffix stripped). Pro / Pro Plus only.

### POST /api/v1/meme/regime/refresh
Force recompute the Meme / Speculative regime cache. Pro / Pro Plus only.

### GET /api/v1/event/regime
Forward catalyst calendar for the Event / Catalyst regime (Regime #5) — the only event-centric regime (discrete, dated catalysts on a forward time axis rather than continuous per-symbol price features). Returns every catalyst in the next `window_days` (default 7, max 30): token unlocks, scheduled macro prints (FOMC/CPI/NFP/PCE/PPI), and stablecoin depegs.
- Each catalyst: `{ event_type (unlock|macro_print|depeg), title, date, days_until, symbols, bias, magnitude, ... }`. Unlocks bias short (supply overhang) scaled by %-of-FDV × proximity; macro prints are always `risk_flag` pre-event (the surprise is unknown ahead of time, so they raise risk without a directional tilt); major depegs bias short scaled by deviation beyond a 25bps dead-zone.
- Also returns the market-wide `event_risk_score` (0-100) and `band`. Unlocks (DefiLlama emissions), macro (FRED + FOMC calendar), and depeg (CoinGecko prices) are backfillable; response also flags `partial` when a feed is down.
- Response: `{ items: [...], count, window_days, event_risk_score, band, timestamp }`. Cached 30 min; refreshed every 30 min by the scheduler.

### GET /api/v1/event/regime/score
Market-wide Event Risk composite (0-100, baseline 0 — event risk is absence-by-default). Bands: dormant (<10), quiet (10-29), elevated (30-54), heavy (55-79), critical (≥80). Composite = 0.40·unlock_subscore + 0.25·macro_subscore + 0.35·depeg_subscore, re-normalized over whichever feeds are available (`partial` / `inputs_available` flag which were live, so a down feed doesn't read as "no risk"). Also returns `events_7d` / `events_30d` counts. A live-only `sector_rotation` sidecar (CoinGecko category leaders/laggards by 24h market-cap flow) sits alongside (NOT inside) the composite, so the live and backfilled scores are computed by the same formula; it is null on backfilled days.

### GET /api/v1/event/calendar
The queryable form of the forward calendar (up to 30d out). Query params: `type` (unlock | macro_print | depeg), `window_days` (1-30, default 30), `symbol` (filter to catalysts affecting a ticker), `bias` (long | short | neutral | risk_flag), `min_magnitude` (0.0-1.0). Same item shape as /event/regime.

### GET /api/v1/event/regime/{symbol}
Per-symbol catalyst overlay — the pending catalysts affecting one coin in the next 7d, its `net_bias` (magnitude-weighted vote across its events), `net_magnitude`, `nearest_event_days`, and `hl_symbol` (the exact Hyperliquid perp join key; null for stablecoins). Path param: bare ticker (USDT suffix stripped). Pro / Pro Plus only.

### POST /api/v1/event/regime/refresh
Force recompute the Event / Catalyst regime cache. Pro / Pro Plus only.

### GET /api/v1/security/regime
Recent security events for the Security / Black Swan regime (Regime #11) — an event-driven regime measuring acute security stress (the forensic signature of exchange hacks, protocol exploits, and stablecoin depeg cascades: immediate OI unwind + self-custody withdrawal flight, sector contagion, then institutional reaccumulation). Returns every security event in the last `window_days` (default 10, max 10): confirmed hacks/exploits (DefiLlama hacks registry) and live stablecoin depegs.
- Each event: `{ event_type (hack|depeg), title, date, days_ago, symbols, bias, severity, amount_usd, classification, technique, target_type, chains, ... }`. Hacks bias short scaled by USD lost × recency; depegs bias short scaled by deviation beyond a 25bps dead-zone.
- Also returns the market-wide `security_stress_score` (0-100) and `band`. Hack / flow / depeg subscores are backfillable; response flags `partial` when a feed is down.
- Response: `{ items: [...], count, window_days, security_stress_score, band, timestamp }`. Cached 30 min; refreshed every 30 min by the scheduler.

### GET /api/v1/security/regime/score
Market-wide Security Stress composite (0-100, baseline 0 — security stress is absence-by-default). Bands: dormant (<10), quiet (10-29), elevated (30-54), heavy (55-79), critical (≥80). Composite = 0.45·hack_subscore + 0.30·flow_subscore + 0.25·depeg_subscore, re-normalized over available feeds (`partial` / `inputs_available`). flow_subscore is the worst absolute net-CEX-flow z-score across the major/stable universe (self-custody-flight signature, either direction is stress); hack_subscore saturates USD-lost × recency over confirmed hacks within 10d; depeg reuses the #5 cascade measure. Also returns `hacks_in_window`, `largest_hack_usd`, `worst_flow_symbol` / `worst_flow_zscore`. A live-only `security_headlines` sidecar (classified rekt.news / SlowMist RSS advisories — `advisory_pressure`, top headlines) sits alongside (NOT inside) the composite; null on backfilled days.

### GET /api/v1/security/events
The queryable form of the recent security-events list (up to 10d back). Query params: `type` (hack | depeg), `window_days` (1-10, default 10), `symbol` (filter to events affecting a ticker), `min_severity` (0.0-1.0). Same item shape as /security/regime.

### GET /api/v1/security/regime/{symbol}
Per-symbol security overlay — the security events implicating one coin, its `acute_bias`, `max_severity`, and `hl_symbol` (exact Hyperliquid perp join key). Path param: bare ticker (USDT suffix stripped). Pro / Pro Plus only.

### POST /api/v1/security/regime/refresh
Force recompute the Security / Black Swan regime cache. Pro / Pro Plus only.

### GET /api/v1/policy/regime
Geopolitical / Policy Shock regime (Regime #12) — a market-wide, event-driven regime measuring policy + geopolitical pressure on crypto (pro-crypto executive orders that trigger euphoric OI build, tariff/trade-war escalations that trigger risk-off macro cascades, central-bank rate signals, country bans/adoptions). Returns the `policy_risk_score`, the signed `policy_tilt` (−1 restrictive/risk-off … +1 pro-crypto/risk-on), `tilt_label`, and the upcoming rate catalysts (scheduled FOMC decisions) within `window_days` (default 45). Unlike the symbol-centric regimes there is no per-symbol overlay — policy moves the whole market. Response: `{ items: [...], count, window_days, policy_risk_score, band, policy_tilt, tilt_label, timestamp }`. Cached 30 min.

### GET /api/v1/policy/regime/score
Market-wide Policy Risk composite (0-100, baseline 0). Bands: dormant (<10), quiet (10-29), elevated (30-54), heavy (55-79), critical (≥80). Composite = 0.40·gdelt_subscore + 0.35·cross_asset_subscore + 0.25·rate_subscore, re-normalized over available feeds. gdelt_subscore is a crypto-policy/geopolitical news-volume z-score (GDELT DOC API — a coverage spike = a shock breaking); cross_asset_subscore is safe-haven dislocation (gold + treasury-yield daily-change z-scores, the tariff-cascade flight-to-safety signature); rate_subscore is proximity to the nearest scheduled FOMC decision. The signed `policy_tilt` comes from GDELT coverage tone (all backfillable). A live-only `policy_headlines` sidecar (Federal Register / SEC / CFTC, classified pro-crypto vs restrictive) sits alongside (NOT inside) the composite; it is null on backfilled days.

### GET /api/v1/policy/headlines
The live breaking-news sidecar — recent US-regulatory headlines (Federal Register documents API + SEC / CFTC press feeds) classified pro-crypto vs restrictive, with an aggregate `regulatory_pressure` (0-1 magnitude), a signed `headline_tilt`, and the top headlines. Live-only — not part of any backfillable composite.

### POST /api/v1/policy/regime/refresh
Force recompute the Geopolitical / Policy Shock regime cache. Pro / Pro Plus only.

### GET /api/v1/backtesting/daily-snapshots
List of dates (YYYY-MM-DD UTC) with an archived daily snapshot available. Snapshots are produced at 20:00 UTC daily + on server startup. Each archived snapshot is identical in shape to /api/v1/daily and now includes onchain_health_score with full component breakdown — ideal for time-series regime tagging in backtests.

### GET /api/v1/backtesting/daily-snapshots/{date}
Retrieve one archived daily snapshot. Full payload: coin profiles, health score, derivatives summary, sentiment, macro indicators, all on-chain collector states, the composite onchain_health_score, the full `technical_regime` block including the per-symbol `by_symbol` map (keyed by ticker — ma/bollinger/range_state/rsi frozen point-in-time; `rsi.*_1h` fields are neutral on backfilled days), and the `signum_rgg` directional trend radar with its per-symbol `by_symbol` map (RED/GREY/GREEN color + signed trend score + ADX/DMI + `hl_symbol` join key; `intraday_*` fields are null on backfilled days). 404 when no snapshot exists for the requested date.

### GET /api/v1/backtesting/snapshots
Time-series JSON snapshots for a single `data_type` (e.g. coinglass_oi, coinglass_funding, stablecoins, market_health, fear_greed, dex_trending). Parameters: `data_type` (required), `start` (required, ISO 8601 or unix ms), `end` (optional), `limit` (default 100, max 1000), `universe` (optional; for dex_trending, `hl_perps` filters to the Hyperliquid perp universe). The response is **streamed row-by-row** so even `limit=1000` pulls of large multi-symbol types stay within bounded server memory and begin returning immediately. Pro Plus. The heavy `/backtesting/*` endpoints are served by an isolated container and are rate-limited per API key. **For bulk multi-day history, prefer `/backtesting/archives/download`** (pre-signed links to pre-built daily files) instead of paginating this endpoint.

### GET /api/v1/backtesting/klines
Historical 1-minute OHLCV klines for backtesting. Pro Plus.
- Parameters: `symbol` (required, e.g. BTCUSDT or BTC), `exchange` (binance | hyperliquid, default binance), `start` (required, ISO 8601 or unix ms), `end` (optional, defaults to now), `limit` (default 1000, max 10000), `format` (json | csv)

### GET /api/v1/backtesting/funding
Historical funding rates and open interest. Pro Plus.
- Parameters: `symbol` (required), `exchange` (binance | hyperliquid, default hyperliquid), `start` (required), `end` (optional), `limit` (default 1000, max 10000)

### GET /api/v1/backtesting/liquidations
Historical liquidation events. Pro Plus.
- Parameters: `symbol` (optional; omit for all), `start` (required), `end` (optional), `limit` (default 1000, max 10000)

### GET /api/v1/backtesting/snapshots/types
List all available snapshot data types with row counts and date ranges. Pro Plus.

### GET /api/v1/backtesting/symbols
List all tracked symbols with their available date ranges. Pro Plus.
- Parameters: `exchange` (optional; binance | hyperliquid)

### GET /api/v1/backtesting/status
Backtesting storage statistics — row counts, date coverage and disk usage. Pro Plus.

### GET /api/v1/backtesting/export
Export kline data as a streaming CSV download (up to 100k rows). Pro Plus.
- Parameters: `symbol` (required), `exchange` (binance | hyperliquid, default binance), `start` (required), `end` (optional)

### GET /api/v1/backtesting/archives/index
Compact index of all archived data types, symbols, exchanges and date ranges on object storage — call once to discover what's available, then use /archives/download. Pro Plus.

### GET /api/v1/backtesting/archives
List archived files available for download from cold storage. Use `data_type=daily` for consolidated per-exchange Parquet bundles (recommended for bulk). Pro Plus.
- Parameters: `data_type` (daily | klines | funding | liquidations | snapshots | klines_deep | funding_deep, default klines), `exchange` (optional), `symbol` (optional), `interval` (klines_deep only: 1h/4h/1d), `start`/`end` (optional YYYY-MM-DD, or YYYY-MM for the *_deep tiers, recommended for large types)

### GET /api/v1/backtesting/archives/download
Generate pre-signed download URLs for archived files — the recommended route for bulk multi-day history. Pro Plus.
- Parameters: `start` (required YYYY-MM-DD, or YYYY-MM for *_deep), `data_type` (daily | klines | funding | liquidations | snapshots | klines_deep | funding_deep, default klines), `exchange` (optional), `symbol` (optional), `interval` (klines_deep only: 1h/4h/1d), `end` (optional, defaults to today)
- DEEP cold tiers (monthly, one Parquet per symbol per month, reaching far past the rolling daily files):
  - `klines_deep` — coarse OHLCV (columns time, open, high, low, close, volume, quote_volume=0). Depth by interval: 1d back to HL's 2023 launch, 4h to ~2024, 1h to ~7 months. Deep 1m does NOT exist (HL's REST candleSnapshot serves ~4 days of 1m; HL's S3 archive is L2-book only). Example: `?data_type=klines_deep&exchange=hyperliquid&symbol=BTC&interval=1d&start=2023-01`
  - `funding_deep` — hourly funding_rate (columns time, funding_rate, premium) back to ~2024 per coin from listing. Historical open_interest / mark_price are unavailable from any HL source, so they are absent (only the rolling 5-min `funding` files carry OI/mark, forward from ~2026-03).

---

## Hyperliquid Perpetuals

### GET /api/v1/hyperliquid/meta
Exchange metadata (assets, max leverage, specs).

### GET /api/v1/hyperliquid/prices
All mid prices across Hyperliquid assets.

### GET /api/v1/hyperliquid/funding-rates
Current + historical funding rates for a coin.
- Parameters: `coin` (required), `limit` (default: 30)

### GET /api/v1/hyperliquid/open-interest
Open interest across all Hyperliquid assets.

### GET /api/v1/hyperliquid/candles
OHLCV candles for a coin.
- Parameters: `coin` (required), `interval` (default: "1h"), `limit` (default: 100)

### GET /api/v1/hyperliquid/l2-book
L2 order book snapshot.
- Parameters: `coin` (required)

### GET /api/v1/hyperliquid/summary
All-in-one perp data for a coin (price, funding, OI, recent candles).
- Parameters: `coin` (required)

## Hyperliquid Trader Intelligence

### GET /api/v1/hyperliquid/top-traders
Scored leaderboard of top Hyperliquid traders filtered by performance criteria.

### GET /api/v1/hyperliquid/wallet-positions
Current positions for tracked wallets. For non-tracked addresses, queries Hyperliquid on-demand.
- Parameters: `address` (optional) — wallet address (0x...). Works for ANY Hyperliquid address.

### GET /api/v1/hyperliquid/wallet-signals
Position change signals: entries, exits, size increases/decreases.
- Parameters: `minutes` (default: 10, range 1-1440), `address` (optional)

### GET /api/v1/hyperliquid/trader-profiles
Trade profiles with win rate, PnL, classification, and edges for all tracked wallets.
- Parameters: `address` (optional)

### GET /api/v1/hyperliquid/trader-profile/{address}
On-demand trade profile for ANY Hyperliquid address. Fetches fill history and computes stats.
- Parameters: `days` (default: 30, range 1-90)

### GET /api/v1/hyperliquid/wallet-trades/{address}
Historical trades for any Hyperliquid address with summary statistics.
- Parameters: `days` (default: 30, range 1-90)

### GET /api/v1/hyperliquid/watchlist
List all watchlisted wallet addresses.

### POST /api/v1/hyperliquid/watchlist
Add wallet addresses to the watchlist. Automatically starts tracking for signals. The watchlist is a shared collector-wide poll set, so mutations require Pro / Pro Plus.
- Body: `{"entries": [{"address": "0x...", "label": "optional", "notes": "optional"}]}`

### DELETE /api/v1/hyperliquid/watchlist
Remove wallet addresses from the watchlist. Pro / Pro Plus.
- Body: `{"addresses": ["0x..."]}`

### GET /api/v1/hyperliquid/wallets/search
Search the cached top-trader leaderboard pool for wallets matching arbitrary performance gates (read-only; does not modify the watchlist). Pool refreshes daily at 19:00 UTC. Pro / Pro Plus.
- Parameters: `min_score` (0-100, default 0), `min_win_rate` (0-1), `min_profit_factor`, `min_trades_30d` (default 30; pass 0 to disable), `min_avg_duration_hours`, `exclude_types` (comma-separated, e.g. hft,scalper), `limit` (1-50, default 50)

### GET /api/v1/hyperliquid/copy-signals
Top traders plus their recent position signals in one response — collapses the top-traders → watchlist → signals workflow into a single call. Only returns traders with signals in the window. Pro / Pro Plus.
- Parameters: `min_score` (0-100, default 80), `minutes` (1-1440, default 60), `min_win_rate` (0-1), `min_profit_factor`, `min_trades_30d`, `min_avg_duration_hours`, `exclude_types`

### GET /api/v1/hyperliquid/watchlist/auto
Current auto-managed watchlist configuration and status (which top-trader criteria drive the managed set). Requires API key.

### POST /api/v1/hyperliquid/watchlist/auto
Auto-populate the watchlist from top traders by configurable criteria (min score / win rate / profit factor, max wallets, ...); re-evaluated on the daily refresh. Pro / Pro Plus.
- Body: auto-watchlist criteria object

### DELETE /api/v1/hyperliquid/watchlist/auto
Disable auto-management and remove all auto-managed watchlist entries. Pro / Pro Plus.

### POST /api/v1/hyperliquid/trader-profiles/refresh
Force a synchronous refresh of trade profiles for all tracked wallets (~15-30s). Pro Plus. Normal operation is covered by the daily 19:00 UTC job.

### POST /api/v1/hyperliquid/leaderboard/refresh
Force a synchronous refresh of the Hyperliquid leaderboard cache (~25-30s, ~28 MB). Pro Plus. Use when /copy-signals or /wallets/search return 0 results and the in-memory leaderboard looks empty.

## Webhooks

### GET /api/v1/webhooks
List all registered webhook endpoints.

### POST /api/v1/webhooks
Register a new webhook endpoint with optional address and event filters.
- Body: `{"url": "https://...", "secret": "...", "label": "...", "enabled": true, "addresses": ["0x..."] | null, "events": ["ENTRY","EXIT","INCREASE","DECREASE"] | null}`

### PATCH /api/v1/webhooks/{label}
Update a webhook endpoint by label.

### DELETE /api/v1/webhooks/{label}
Delete a webhook endpoint by label.

---

## Market Data (Binance Spot)

### GET /api/v1/market-data/klines
OHLCV klines from Binance Spot.
- Parameters: `symbol` (default: "BTCUSDT"), `interval` (default: "1d"), `limit` (default: 30)

### GET /api/v1/market-data/ticker/24hr
24hr ticker stats from Binance Spot.
- Parameters: `symbol` (default: "BTCUSDT")

### GET /api/v1/market-data/ticker/price
Current price from Binance Spot.
- Parameters: `symbol` (default: "BTCUSDT")

### GET /api/v1/market-data/btc-price-history
BTC price history with 200-day moving average.

### GET /api/v1/market-data/volume-history
Daily volume + buy ratio from remote source.

### GET /api/v1/market-data/short-term-price
Short-term BTC price momentum metrics.

### GET /api/v1/market-data/exchange-info
Exchange pair info from Binance Spot.

---

## DEX & Meme Coins

### GET /api/v1/dex/trending
Trending DEX pools across chains — high-frequency discovery data for meme coin alpha.
- Parameters: `chain` (optional) — filter by chain: "solana", "ethereum", "base", "bsc", "arbitrum"
- Response: `{ chain, count, pools: [{ chain, dex, pool_address, name, created_at, base_token: { address, name, symbol }, quote_token: { address, name, symbol }, price_usd, fdv_usd, reserve_usd, volume: { m5, h1, h6, h24 }, price_change_pct: { m5, h1, h6, h24 }, transactions: { h1: { buys, sells }, h24: { buys, sells } } }] }`
- Refresh: Every 2 minutes

### GET /api/v1/dex/new-pools
Newest DEX pools — early discovery of new token launches.
- Parameters: `chain` (optional) — filter by chain: "solana", "ethereum", "base", "bsc", "arbitrum"
- Response: Same shape as /dex/trending

### GET /api/v1/dex/token/{chain}/{address}
Token info + top pools for a specific token on a chain.
- Path parameters: `chain` — solana, ethereum, base, bsc, or arbitrum. `address` — token contract address.
- Response: `{ name, symbol, description, chain, token_address, gt_score, metadata, top_pools: [...] }`
- Caching: 5-minute per-token LRU cache

### GET /api/v1/dex/promoted
Recently promoted/boosted tokens — marketing spend signal. Tokens paying for promotion indicates active marketing campaigns.
- Response: `{ count, tokens: [{ chain, token_address, description, promotion_spend, active_promotion, has_profile, links: { website, twitter, ... } }] }`
- Refresh: Every 5 minutes

### GET /api/v1/dex/promoted/top
Top promoted tokens ranked by total promotion spend.
- Response: Same shape as /dex/promoted

### GET /api/v1/dex/security/{chain}/{address}
Token security report — rug detection, honeypot check, risk scoring.
- Path parameters: `chain` — solana, ethereum, base, bsc, or arbitrum. `address` — token contract address.
- Response: `{ chain, token_address, token_name, token_symbol, risk_flags: { is_honeypot, is_mintable, is_freezable, has_hidden_owner, is_open_source, is_blacklistable, can_pause_transfers, has_tax_modification }, taxes: { buy_tax_pct, sell_tax_pct }, ownership: { creator_address, creator_balance_pct, owner_address }, holder_info: { holder_count, top_10_pct }, lp_info: { total_lp_holders, any_lp_locked }, risk_score, risk_level }`
- Risk score: 0-100 (100 = safest). Levels: critical (0-25), high (26-50), medium (51-75), low (76-100)
- Caching: 10-minute per-token LRU cache

---

## Payments

### GET /api/v1/payments/plans
Plan pricing and supported networks (no auth required).

### POST /api/v1/payments/subscribe
Subscribe to a pro plan via x402 payment.

### GET /api/v1/payments/subscription
Current subscription status.

### GET /api/v1/payments/invoices
List all invoices for this API key's email.

### GET /api/v1/payments/invoices/{number}
Single invoice (JSON, HTML, or PDF format).

### POST /api/v1/payments/agent-subscribe
Gasless one-shot onboarding for AI agents (x402 payment protocol).

---

## Authentication

### POST /api/v1/auth/keys
Create a new free-tier API key.

### GET /api/v1/auth/keys/me
Get current API key info.

### POST /api/v1/auth/keys/rotate
Rotate (invalidate old, create new) API key.

### DELETE /api/v1/auth/keys
Revoke the current API key.

---

## Wallet Authentication

### POST /api/v1/wallet/challenge
Generate an EIP-191 sign-in challenge.

### POST /api/v1/wallet/verify
Verify an EIP-191 signature and return/create an API key.

### GET /api/v1/wallet/session
Get wallet session info.

### POST /api/v1/wallet/upgrade
Verify a USDC transfer on-chain and upgrade to pro.

### GET /api/v1/wallet/invoices
List invoices for current wallet.

### GET /api/v1/wallet/invoices/{number}
Download a single invoice by its number (PDF).

### GET /api/v1/wallet/config
Public wallet config (treasury addresses, networks).

### POST /api/v1/wallet/create-solana-tx
Build a Solana SPL USDC transfer transaction.

### GET /api/v1/wallet/solana-blockhash
Get latest Solana blockhash.

### GET /api/v1/wallet/solana-confirm/{signature}
Check Solana transaction confirmation status.

---

## Health Check

### GET /api/v1/health
Basic health check (no auth required).
- Response: `{ status: "ok" }`