f1ow MCP Server

OHLC(V) candle history for a coin by symbol (e.g. 'ETH', 'BTC'). 'interval' picks the candle size (1m|5m|15m|30m|1h|4h|12h|1d|1w, default 1d); 'limit' the number of candles (default 100, max 400); 'market' picks the series — 'spot' (Binance, Bybit, then CoinGecko/CoinMarketCap), 'perp' (Hyperliquid, Bybit, then Binance futures — perpetual marks), or 'auto' (default: spot preferred, perp fallback). spot/perp are hard constraints and error rather than substitute. The response labels the market, interval and source actually delivered, includes USD volume when the source carries it, and flags a still-forming last candle.

Fetches price, market cap, ATH/ATL and description for a coin by its coin ID (e.g. 'bitcoin', 'ethereum')

Lists the coin symbols available from the market-data provider

Global market snapshot: total market cap, BTC/ETH dominance, Fear & Greed index, and Altcoin Season index

Per-coin market snapshot (spot price, perpetual mark price, 24h change, funding rate, open interest, long/short ratio, 24h liquidations), merged across providers. Spot and perp prices are separate fields (spot from spot-market providers, perp mark from the perp venue); they differ by the basis. Pass 'symbols' to filter to specific coins; omit it for every covered coin.

Ranked 24h gainers, losers, and trend candidates (>+5% move)

Lists the currently trending cryptocurrencies

Cross-sectional opportunity scan across every major and medium-cap coin at once — the screener the per-coin tools cannot be looped into safely. Two whole-universe feeds (the merged market snapshot + venue perp contexts) build a liquidity-filtered universe and rank it, then candles are fetched for a bounded shortlist only; ZERO per-symbol calls are made to the rate-limited derivatives provider, which is what lets this cover a universe without silently degrading. Up to six signals per candidate: crowding (annualized funding vs the venue-standard anchor AND its percentile across the screened universe right now — a cross-sectional read, explicitly NOT the own-history percentile research_positioning reports), stress (24h liquidations as a share of open interest — whether forced sellers already cleared), leverage (open interest as a share of market cap — the gap-through-your-stop condition), compression (4h Bollinger bandwidth ranked against the coin's own trailing window — the per-coin form of the reading research_regime computes for BTC alone; says energy is loading, never direction), trend (ADX with the +DI/-DI separation on 4h, confirmed against 1d — the vote is withheld unless the slow timeframe agrees), and location (distance from the 4h EMA20 in ATR units — an extended coin scores DOWN, because a chase is not a setup). Each signal reports its state, the numbers, and a +1/0/-1 setup-quality vote; the setup name and score come from a disclosed rule shipped with every scan. Candidates carry ATR-and-structure geometry — a resting-limit entry at the EMA20 (never at market), a stop a full 2 ATR beyond it, and a target shaded inside the first structural level — priced so it feeds risk_position_size directly. 'tier' filters major/medium/all; 'limit' caps the ranked list; 'symbols' narrows the universe to specific coins, and every symbol named there is accounted for individually in the response — selected, excluded by your own tier filter, below the liquidity floors (never screened, so no band was ever applied to it), below the tier floor, or not quoted at all — because mere absence from the ranked list conflates four different answers. Reports a board, not a forecast: the bands are disclosed constants chosen for plausibility with no backtest behind them, and 'NO CANDIDATES' is a real and frequent answer. Use it to pick what to look at, then research_positioning and ta_technicals on the two or three that stand out.

Deterministic positioning read for one coin — the game-state companion to research_regime, answering 'who is crowded, who is paying, who is trapped, and where'. Up to nine axes, each tagged with the market-actor cohort it reads: crowding (annualized funding vs the venue-standard 0.01%/8h anchor, long/short skew, dollars/day the majority pays to hold, crowd-vs-top-trader divergence, and cross-venue funding dispersion), buildup (24h price vs open-interest direction and the funding trend — whether positions are being added into the move or closed out), liquidity (order-book spread and imbalance — makers backing off), basis (perp vs oracle premium — spot-perp froth or hedging pressure, informational), fragility (24h liquidations as a share of market-wide open interest; the axis degrades rather than substituting a single venue's OI, since the liquidation total is cross-exchange), predation (visible Hyperliquid whale positions within 2 daily sigmas of their liquidation price, with defending book depth), trap (share of tracked whale notional underwater on the funding-paying side — attrition fuel), house book (Hyperliquid HLP LP-vault inventory, summed across its child sub-vaults — the literal venue counterparty, so the inventory it carries mirrors how traders are crowded; measured as net over its own gross book, since a market maker's net is a rounding error against venue open interest; informational), and disagreement (the most liquid matching prediction market, volume-gated, informational). Each axis reports its state, the numbers, and a +1/0/-1 fragility vote (+1 clean, -1 crowded/fragile); the stance is the disclosed sum-of-votes rule. Also returns a one-line 'farmed' synthesis (which cohort the board is currently farming, or an explicit statement that none is), focal points — reachable liquidation and breakeven price levels the whole market can see — and 'would change the call' thresholds a polling agent can watch statelessly. Reports the board, not a direction: crowding says who is paying, not where price goes. Deterministic — same inputs, same read. Call research_regime first for market context, then this per coin before sizing or timing decisions.

Deterministic market-regime classification across seven axes: trend (Mayer Multiple bands vs the 200-day SMA), volatility (30d realized, percentile vs own history), cycle (a ~30-indicator top checklist incl. MVRV Z-Score, NUPL, Pi Cycle, AHR999), sentiment (Fear & Greed level, trajectory and percentile), leverage stress (derivatives-risk-index percentile), liquidity (7d spot-ETF net flows vs prior 7d), and rotation (Altcoin Season index, BTC dominance, ETH/BTC 30d). Each axis reports its state, the numbers behind it, and a +1/0/-1 vote; the overall posture is the disclosed sum-of-votes rule, with one exception — while 30d realized volatility sits at or below its 5th percentile the tape is compressed, trend structure stops being tradeable, and the Trend axis's vote is withheld (its reading is still reported). Compression is flagged structurally, and adds an informational, never-voting asymmetry lean naming which side is more exposed. Includes explicit 'would change the call' thresholds to watch. Rule-based and deterministic — same inputs, same read. Useful as a first call: most other signals (funding, sentiment, flows) read differently depending on this regime.

Aggregated single-coin dossier: market data (spot/perp price, funding, market-wide OI, long/short, liquidations — the same enriched line as market_quotes), global market context (dominance, Fear & Greed), multi-timeframe technicals computed in-house from candles (4h + 1d: RSI, EMA/SMA, MACD, ATR, Bollinger, ADX, trend — the same numbers as ta_technicals), on-chain TVL/fees (when the symbol is a tracked chain), news, a grounded narrative synthesis with citations, upcoming catalyst events, social sentiment, ETF flows, prediction-market odds, live provider coverage, and a per-section status list that distinguishes an empty section from a failed feed — for one symbol (e.g. 'ETH').

Lists upcoming crypto catalyst events for the next 7 days. Optionally filter by comma-separated coin IDs (e.g. 'bitcoin,ethereum').

Scheduled macroeconomic data releases (CPI, FOMC, jobs, GDP, etc.) from the last ~2 days through the next 7 days. Optionally set 'min_importance' (1-3, default 2) to filter out low-impact releases.

Recent daily spot-ETF net flows (USD) for an underlying asset (BTC, ETH, SOL, XRP; defaults to BTC): latest day total, per-fund breakdown, and the trailing daily trend. Positive means net inflows. Flows settle behind spot and skip weekends, so the latest row is routinely a day or more old — its age is reported next to the date, and the price it carries is the price on THAT date, not spot. 'symbol' is accepted as an alias for 'asset'.

Order-book snapshot for one coin's perpetuals: best bid/ask, spread in bps, top-of-book depth (USD) per side, and bid/ask depth imbalance. Positive imbalance means more bid depth. Single-venue by nature — Hyperliquid, falling back to Bybit for coins it doesn't list; the response names the venue that answered.

Aggregate whale context for one coin (cohort-level, no individual wallets): per-exchange reserve changes (1d/7d), top-trader long/short positioning on Binance perps, and large on-chain transfers to/from exchange wallets above a USD threshold. The transfer feed covers ERC-20 tokens; for native assets (BTC, ETH) it substitutes market-wide USDT exchange flows as a dry-powder signal. Exchange withdrawals suggest accumulation; deposits suggest potential sell pressure. For specific whale wallets and their live positions use flow_whale_positions.

Live whale positions on Hyperliquid perps. By default discovers the largest currently-active whale accounts from the venue leaderboard and returns each one's equity and open positions with side, notional (USD), entry price, liquidation price, leverage and unrealized PnL. Optionally pass 'addresses' (comma-separated 0x wallets, max 10) to follow specific wallets instead. For aggregate cohort signals use flow_whale_context.

Current market-wide Fear & Greed index (0-100 plus label), updated roughly every 15 minutes. A macro sentiment gauge; use market_overview for the full market snapshot.

Social-sentiment snapshot for one coin: galaxy score (0-100 composite), alt rank (lower is stronger), bullish sentiment %, 24h social volume, social dominance %, and 24h interactions.

Searches Twitter/X and returns matching tweets with engagement (likes, retweets, replies, views) and author reach (followers, verified). Raw crowd voice for judging social sentiment on a coin or topic. Sorted by relevance; set latest=true for the newest tweets instead.

Current market-wide (cross-exchange aggregate) open interest (USD) for one coin's perpetuals, with its 24h and 7d change, the 24h price change, AND the underlying 4h-bucket series. Read OI against price: price up on rising OI is new money opening positions, price up on falling OI is short covering — opposite trades off the same price move. The series makes that read available bucket by bucket instead of once per day (pair it with market_candles at 4h for the price leg), and carries the window's own low/high so the current level can be judged against where it has actually been. Falls back to a single-venue figure if the aggregate source is offline. Set 'points' to trim the series (omit for the full ~10-day window; 0 to suppress it).

Market-implied prediction-market odds for a topic (e.g. 'Fed rate cut', 'bitcoin 100k'). Returns the most active matching markets with per-outcome probabilities, volume, and resolution date.

Searches recent regulatory filings (full-text) for a company, ticker, or theme (e.g. 'bitcoin treasury'). Returns the most recent matching filings with form type, filer, and a direct document link.

Current perpetual funding rate for one coin (hourly rate plus an annualized estimate), merged across providers, with its 24h and 7d trajectory AND the underlying 8h-bucket series. Positive means longs pay shorts. The trajectory is the point: the same +6% annualized reads as post-flush relief if it fell from +40% and as shorts capitulating if it rose from -20%. The series carries the shape a trailing delta flattens away — a steady drift and a spike that retraced give the same 24h change off opposite tapes — plus the window's own low/high, so the current rate can be judged against where it has actually been rather than an absolute threshold. Set 'points' to trim the series (omit for the full ~10-day window; 0 to suppress it).

Searches the web for news and articles on a topic from the last 7 days (web results recency-filtered; crypto-native headlines merged in when that feed is up)

On-chain TVL and network fees/revenue (24h/7d/30d) for one chain. Use a chain's own token symbol (e.g. 'ETH', 'SOL', 'BNB') — returns a plain 'not a tracked chain' message for tokens that aren't L1/L2s. Does not cover gas price (gwei).

Reports the health of each connected data provider

Fixed-fractional position sizing — a local calculation, not a data-provider lookup. Given account_size, risk_percent, entry and stop, returns position size in units, notional, implied leverage, stop distance, and — if a target is given — the reward:risk ratio. Direction (long/short) is inferred from the stop's side of entry. Deterministic: same inputs always give the same result.

Multi-timeframe technical-analysis snapshot for one coin, computed locally from real candles (not a vendor black box): per interval — last close, RSI(14), EMA20/50/200 + SMA200, MACD(12,26,9) with signal/histogram/cross freshness, ATR(14) with 2xATR stop suggestions (feed risk_position_size), Bollinger(20,2) with %B and bandwidth, ADX(14) with +DI/-DI, volume vs 20-candle average, and a transparent trend rule (EMA alignment qualified by ADX). 'intervals' picks the timeframes (default 1h,4h,1d; max 4 of 1m|5m|15m|30m|1h|4h|12h|1d|1w); 'market' routes candles (spot|perp|auto, default auto: spot preferred, perp fallback). AltFINS vendor support/resistance levels are appended when that feed is up.

Returns the current UTC time in ISO 8601 format. Use it to anchor 'now' — e.g. to judge how stale a generatedAt timestamp or news date is.