Start here when unsure. Returns the map of flowscan.xyz pages -> tools, what Flowscan does NOT serve (block/tx lookups, prices), the mainnet-only rule and output-shaping conventions. Optionally filter by a keyword (e.g. 'revenue', 'address', 'hip-3'): matching pages plus matching notServed items, with servedByThisServer=false when only notServed matches. Also lists HIP-3 dex display names vs on-chain prefixes.

Homepage 'Stablecoin Perp Margin': total stablecoin value on HyperCore split into spot balances and perp margin (with unique holders/traders), and per token (USDC, USDT, USDE, USDH) spot/perp balances, holders, average/median, total value and market share %. USD.

The /peers gossip-network crawl (~600 nodes). Default: meta (crawl time, counts, reachability, states), footprint (top countries/ASNs) and sentries (validator sentries with operator, state, peers served). section='nodes' (IP id, role, operator, state, tier, hops, parent, feeders, geo, ASN; filterable), 'edges' (feed links) or 'all' are paged.

The /validators (Staking) page: total HYPE staked, delegator count, validator count, and every validator (~35) with name, address, commission (bps), total delegated HYPE, staker count and jailed flag (descriptions only with includeDescription). Filter by name/address, drop jailed validators, sort ascending or descending: e.g. sortBy=commission_bps, order=asc, excludeJailed=true finds the cheapest active validator.

Staking page validator drill-down: one validator's summary (name, commission, total delegated, staker count, jailed) plus its delegators (stakers) as {address, amount HYPE}, largest first. validator is an address or a validator name (e.g. 'Hyper Foundation 2'). Large validators have thousands of stakers; use limit/offset or search.

Staking page validator activity: most recent delegation/undelegation events for one validator, newest first (user, amount in HYPE, isUndelegate, tx hash, time ms + ISO). validator is an address or a validator name. The upstream currency field reads 'USDC' but staking amounts are HYPE.

/revenue 'Daily HyperCore Revenue' (and homepage 24h panel): one row per UTC day, oldest first: nativeHypercoreFee (non-HIP-3 markets) and hip3HypercoreFee (HIP-3 markets), USDC strings, plus rangeTotals. days = last N COMPLETE UTC days ending yesterday (default 90), like revenue_summary and the builder tools; includeToday appends today's partial row (partial: true). Or pass startDate/endDate or startTime/endTime (ms). History from 2026-03.

/revenue 'Deployer Fees': HIP-3 deployer fees per UTC day, oldest first: totalFee and byDex [{dex, totalFee}] with on-chain names ('xyz', 'para', 'io', 'mkts', 'hyna', 'cash', 'flx', 'vntl', 'km', 'hyperliquid'), plus rangeTotals (overall and per DEX). days = last N complete UTC days ending yesterday (default 90); includeToday adds today's partial row. USDC. Paid to deployers, so not part of Flowscan's headline protocol revenue.

/revenue 'Daily Priority Gas' and 'Top Users': per UTC day, writePriority and readPriority {totalGas (HYPE), count} and, with includeTopUsers, the top 5 gas payers, plus rangeTotals. days = last N complete UTC days ending yesterday (default 90); includeToday adds today's partial row.

Convenience aggregate of the /revenue page and the homepage '24h Revenue' card: sums native HyperCore fees (USDC), HIP-3 HyperCore fees (USDC), their sum totalUsdcExcludingGas, HIP-3 deployer fees (USDC, paid to deployers, not protocol revenue) and priority gas (HYPE) over the last 1, 7 and 30 complete UTC days, plus the current (partial) UTC day separately and an annualized run-rate from the trailing 7 complete days. Computed from the three /api/gossip/info series on flowscan.xyz.

Homepage perp snapshot for every perp market (~330 incl. HIP-3 like 'xyz:TSLA'): long/short counts and notional, ratios, open interest, avg entry, median leverage, unique addresses, snapshotIso/snapshotAgeSeconds (refreshes every few minutes). openInterest is Flowscan's two-sided figure: long + short notional (2x the one-sided OI some UIs show). Prefer this for live OI/positioning; hip3_markets/hip3_dex give HIP-3 history, hip3_binance_comparison the Binance side.

Perp snapshot drill-down: open positions in one market, largest first (address, signed size, notional, side, entry, leverage, liq price, account value, funding/all-time PnL), with snapshotIso/snapshotAgeSeconds, marketSummary (whole market, unfiltered; OI = long + short notional) and filteredSideSummary (over the filtered rows only, with filter). Market is case-insensitive; bare 'TSLA' resolves to 'xyz:TSLA' when unique.

Homepage perp snapshot address lookup: an account's open perp positions across all markets incl. HIP-3 (market, side, signed size, notional, entry, leverage, liquidation price, funding PnL, all-time PnL) from the latest snapshot (refreshes every few minutes), with totalNotional and totalPositions. Sorted by notional desc. For live margin/account state use flowscan_address_summary.

Overview of an /address page: role (user/vault/subAccount/agent/missing), lifetime PnL summary (PnL, win rate, trades, hold time, volume, fees, funding, days active, tradedPairs as count + first 30), live perp state (account value, notional, margin, withdrawable, positions with size/entry/leverage/liquidation/uPnL/ROE/funding, largest first, capped by positionsLimit) and non-zero spot balances. Perp state is the main DEX unless dex names a HIP-3 DEX (prefix or display name; unknown names are rejected).

Address 'Open Orders' / 'Order History'. kind='open' (default): all resting orders on every DEX (coin, side B/A, limit price, size, oid, time). 'openDetailed': main-DEX orders with trigger/TP-SL/reduce-only/type/TIF (upstream max 100, capped then). 'historical': the newest ~2000 orders with final status, newest first; countsByStatus (filled/canceled/...) covers all of them, coveredRange shows their time span (seconds for busy accounts). Returns count of matching orders; pages default to 100 (open) / 50 rows. Times have ISO twins.

Address 'Trades' tab: fills newest first, 50/page (coin, px, sz, side, direction, closed PnL, fee, hash, time + timeIso). totals (count, closedPnlUsdc, fees, volumeUsd = px*sz, byCoin) cover ALL matched fills in coveredRange: quote them, never add rows. No startTime: the latest ~2000 fills. With startTime[/endTime]: pages past the 2000-row cap are followed (maxPages); if still capped, continue from nextStartTime.

Address 'Funding' / 'Ledger' tabs, newest first, rows with timeIso. kind='ledger' (default, 50/page, since 30 days ago): deposits, withdrawals, sends, transfers, vault/staking moves; totals.byType {count, sumUsdc, inUsdc, outUsdc}. kind='funding' (100/page, last 7 days): hourly payments; totals {netUsdc, paidUsdc, receivedUsdc, byCoin}. Totals cover ALL matched rows in coveredRange: quote them, never add rows. Pages past the 2000-row cap are followed (maxPages); if still capped, continue from nextStartTime.

Address page 'Staking' section: totalDelegatedHype, current delegations per validator (validator address and name, commission_bps, is_jailed, amountHype, lock-up end) and the staking history (delegate/undelegate, deposits/withdrawals to staking, newest first).

Address 'Vaults' and 'Sub-accounts': equity held in each vault (vault, equity, lock-up) and sub-accounts (name, address, account value, notional, withdrawable, open positions, non-zero spot balances).

Smaller address-page widgets. kind: 'approvedBuilders' (builder codes the user approved with max fee), 'borrowLend' (HyperCore native borrow/lend state per token, health and health factor), 'rateLimit' (API request allowance vs cumulative volume), 'twapSliceFills' (fills generated by the user's TWAP orders).

The /spot-stocks page: tokenized stocks on Hyperliquid SPOT: NVDAX, SPYX, QQQX, SKHYX, MUX, SNDKX, SPCXX, TSLAX, AAPLX, CRCLX (xStocks) and SPCXD (Dinari). Per-token price, 24h/all-time volume, holders and liquidity ARE served. section='current' (default): summary + per-token stats. 'timeseries': daily volume/holders/traders/value per token (30 days default). 'liquidity': cumulative depth within 2/5/10/25 bps in tokens and USD. 'topHolders': largest holders.

The /weekend-trading selector: tracked weekends, newest first (fridayCloseTs = week id, sundayCloseTs, ISO twins, status, avg % change, risers/fallers, market count, data availability). includeSchedule adds the next ~12 weekend/holiday sessions.

Weekend page price table: for HIP-3 TradFi perps (e.g. 'xyz:TSLA', 'xyz:GOLD'; stocks, indices, commodities, FX), the Friday-close prices, Sunday-close prices (once the weekend is over), live current prices (during an active weekend), and weekendChanges {pct, dollar} per market, plus status and price counts.

Weekend page positioning panel: per perp market (crypto and HIP-3 TradFi, ~330), long/short counts and notional at Friday close vs the latest snapshot, new/closed longs and shorts, net long/short changes, and optionally the top 5 address-level position changes per market.

Weekend page coin drill-down: one HIP-3 TradFi market's Friday-close -> Sunday-close move (pct and dollar) for every tracked weekend, newest first. Only DEX-prefixed TradFi markets have data; a bare symbol like 'TSLA' is resolved to 'xyz:TSLA' (crypto like 'BTC' has no weekend data). Timestamps come with ISO twins.

The /hip-4 page: HIP-4 prediction markets. section='active' (default): slim rows (outcomeId, name, marketType, asset ids, underlying/target/expiry, yesMark/noMark = implied probability, yesChange24h, volume24h, totalVolume, deployer, question) sorted by sortBy (default volume24h desc). 'settled': resolved outcomes with settleFraction and trade stats. 'questions': question groups. 'all': all three. full=true for descriptions. Candles: flowscan_hip4_outcome.

HIP-4 page outcome drill-down: candles for the YES and NO assets of one outcome over the last N days (compact rows [openTime ms, open, high, low, close, volume, trades]), plus per-side trade stats and, for settled outcomes (settled=true), the settled outcome record. Settled outcomes typically have no candles. Only outcomeId is required (asset ids default to '#0' YES / '#1' NO). Get outcomeId from flowscan_hip4_markets.

Resolve HIP-4 outcome asset ids like '#14730' (YES) / '#14731' (NO) to readable labels such as 'Arsenal · Yes', as the address page does for HIP-4 spot balances. Returns {labels: {assetId: label}}.

The /hip-3 headline: HIP-3 totals (volume all-time/30d/90d, trades, traders, new users, OI), per-DEX and per-collateral market share, collateral per DEX, builder-routed share (top 3 per DEX), and dexAliases (display name -> on-chain prefix, e.g. KM=mkts, Paragon=para, Entropy=io). OI is two-sided (long + short). Prefer this for DEX-level totals and share.

HIP-3 page charts: daily values per DEX as dated rows [{date, XYZ: v, KM: v, ...}] (raw=true for positional {dates, series}). metric: volume, trades, traders, new_users, oi, oi_by_market (top markets), collateral_traders, collateral_oi. Returns the last N days (default 30); lastDayPartial flags when the last date is the current, still-accumulating UTC day. dex takes a display name or on-chain prefix.

HIP-3 market tables. Without symbol: all HIP-3 markets (dex, symbol, canonical, asset class) filterable by dex/assetClass/search. With symbol ('TSLA', 'GOLD'): per listing DEX the OI, DAU, volume, spread, slippage (1k-1m), plus OI/DAU history trimmed to days. Prefer this to compare one underlying across HIP-3 DEXs; live positioning is in flowscan_perp_markets ('xyz:TSLA'), Binance in flowscan_hip3_binance_comparison.

HIP-3 per-DEX tab: collateral, total volume/trades/traders/OI, every market with all-time volume, traders and current two-sided OI (sorted by volume), and dated daily-total rows for the last N days (today flagged partial). dex: display name or prefix ('mkts' = KM). Prefer this for one DEX's market list.

HIP-3 'Builders' section: share of HIP-3 volume routed via builder codes (all-time/30d/90d), per-DEX builder volume with top builders, and 800+ builders ranked by HIP-3 volume only (address, name, total/30d/90d volume, shares). dex ranks by one DEX. For revenue/users across Hyperliquid use flowscan_builders_leaderboard.

HIP-3 page Binance comparison: ~240 real-world-asset symbols (equities incl. HK/KR/CN, commodities, indices, FX, pre-market) with Binance USDT-M symbol, open interest (USD), 24h volume (USD) and last price. With symbol: Binance row, daily Binance OI/volume history and the HIP-3 side per DEX (OI, DAU, volume, spread, slippage). Prefer this when comparing HIP-3 with Binance.

The /builders 'Builder Arena': ~1800 builders ranked by one metric over a FIXED window (1d/7d/30d/90d/all_time): revenue, volume (USD), new_users, total_users or avg_revenue_per_user_all_time (last two all-time only). Compact rows (rank, id, name, category, value, that metric's windows, all-time revenue, users); full=true for all metrics. For arbitrary ranges like 'last 45 days' use flowscan_builder_revenue; flowscan_hip3_builders ranks by HIP-3 volume only.

The /builders page headline totals: all-time builder revenue (USD), volume (USD), users and avg revenue per user, overall and per builder category (with builder counts), plus the per-builder all-time list (paged with limit/offset, default 50).

The /builders daily revenue chart: for each UTC day in the range, builder revenue (USD) keyed by builder id (slug like 'phantom' for well-known builders, 0x address for the rest), plus range totals per builder and the grand total. Defaults to the last 30 days ending yesterday, like the site. By default keeps the top 20 builders by range revenue. For one builder's total over a range use flowscan_builder_revenue instead.

Resolve a builder name to its id/address. Names can be ambiguous (e.g. two 'fomo' builders); call this first, then pass the exact id/address to flowscan_builder_revenue or flowscan_builder_dashboard (as address or 'id:'). If multiple strong matches exist, show them to the user. Matches id, name and address (case-insensitive); exact matches first, then substrings, by all-time revenue. Each match: id, name, category, address, revenue and volume USD for 1d/7d/30d/90d/all_time, total_users.

Total and daily revenue (USD) of ONE builder over any range ('past 45 days'): days (complete UTC days ending yesterday, default 30) or startDate/endDate (end clamped to yesterday; future start = error). builder: 0x address (preferred), 'id:' or a name; ambiguous names return candidates (ask the user). Sums /api/builders/all-daily-revenue, cross-checked with the dashboard series when the address is known (both totals + covered dates, volume/fills/traders). Daily rows newest first.

The /builders user-growth chart: daily count of active traders and new traders across all builders and per builder since 2025-07-27. Prefer this for daily active/new trader time series; for windowed totals use flowscan_builders_leaderboard (its new_users.7d is a separate dataset and can differ).

The /builders/{id} dashboard for one builder over a FIXED window (7d/30d/90d/all, ending yesterday UTC): stats (volume, revenue, fills, unique/new traders, volume per trader, share of Hyperliquid and builder volume, revenue run-rate), daily series (last 90 days unless days) and volume/revenue by asset. USD. builder: 0x address, 'id:' or a name. For arbitrary ranges use flowscan_builder_revenue.

The /builder-intelligence index: ~120 analysed builders (id, name, category, total/active users, all-time revenue/volume USD, 7d new users) and categories. Ids feed flowscan_builder_intelligence_detail, category ids flowscan_builder_intelligence_summary. Prefer this for user-status/retention questions. Its '7d new users' comes from a different dataset than the leaderboard's and user_series'; say which you quote.

A /builder-intelligence builder report: user status (active/dormant/cooling-off/switched/moved-on), revenue, cohorts, lifecycle, retention, daily activity, top users, heatmap, daily revenue. Payload is ~11 MB, so pick sections (default metadata, key_metrics, user_status_metrics, revenue_metrics). Long lists are paged with limit/offset (50 top-level, 20 nested) and address lists become {count, sample}.

The /builder-intelligence aggregate view for a category (e.g. wallet, copytrading, desktop_trading, mobile_trading) or 'overall': metadata (builders included and their weights), totals (users, active users, 7d new users, all-time revenue, fees 24h/7d/30d/90d, average daily revenue, week-1/4 retention) and user-weighted averages (key metrics, user status, revenue by status, lifecycle, retention, equity/fee cohorts). Cohort member address lists are reduced to {count, sample}.