Return current HTTP-cache statistics.

Reach for this when:

• The user asks why a price/value looks stale or is identical to a previous query (responses may be served from cache up to the per-endpoint TTL).

• You're debugging rate-limit (HTTP 429) errors and want to confirm the cache is doing its job.

• The user explicitly asks about cache utilization or hit rate.

Returns: Dict with keys: entries: number of live cached entries max_entries: LRU eviction threshold hits, misses, sets, errors: cumulative counters since process start hit_rate: hits / (hits + misses), 0.0 if no requests yet by_pattern: per-endpoint-pattern breakdown of the same counters

Drop every entry from the HTTP cache.

Reach for this when:

• The user explicitly asks to refresh / invalidate cached data.

• You suspect cached data is materially stale and a TTL has not yet expired (e.g. user reports a price that disagrees with their exchange).

Prefer letting TTLs expire naturally; this tool resets ALL endpoints, not just one. Counters (hits/misses/...) are preserved.

Returns: {"cleared": }

Compare the current funding rate for one perp across multiple exchanges, in parallel.

Use this to find funding-rate arbitrage opportunities (long the venue paying you, short the venue charging you) or to gauge how lopsided positioning is across the market. The spread_bps field is (max - min) * 10000 and tells you how big the dispersion is in basis points. For a single exchange snapshot use get_funding_rate; for historical trend on one venue use get_funding_rate_history.

Symbol convention: linear USDT-margined perps use "BTC/USDT:USDT" on most venues. BitMEX's flagship is the inverse contract "BTC/USD:BTC", so if you pass a USDT linear symbol the BitMEX branch will return an error inline — that's expected. Branches that fail (unsupported symbol, geo- block, rate limit) return {"exchange": ..., "error": ...} rather than sinking the whole call.

Args: symbol: CCXT unified perp symbol. Default "BTC/USDT:USDT". exchange_ids: Comma-separated CCXT exchange IDs. Default "binance,okx,bybit,bitmex".

Returns: {"symbol", "rates": [...], "max", "min", "spread_bps", "n_ok", "n_error"}. Each rate row is either {"exchange", "fundingRate", "nextFundingTimestamp", "markPrice"} on success or {"exchange", "error"} on failure. max/min/ spread_bps are populated only when at least two branches returned numeric fundingRate values.

Concurrently fetch the price of one coin from many sources and compare.

Use for cross-source sanity checks, finding venue-vs-aggregator divergences, or simple arbitrage spotting. Contrast with:

• get_price: single source (CoinGecko aggregated).

• compare_funding_rates: perp funding across venues.

Sources fanned out in parallel:

• CoinGecko aggregated /simple/price

• For each exchange_id: CCXT fetch_ticker(base/quote) where base is derived from coin_id via a small mapping (bitcoin->BTC, ethereum->ETH, solana->SOL, ripple->XRP, cardano->ADA, dogecoin->DOGE, tron->TRX, polkadot->DOT, chainlink->LINK, avalanche-2->AVAX, matic-network->MATIC) and quote is USDT for vs_currency=usd, else vs_currency.upper().

• DexScreener (only when vs_currency is "usd"): top-liquidity pair.

Args: coin_id: CoinGecko coin ID (e.g. "bitcoin"). vs_currency: Quote currency. "usd" works on all sources; others skip DexScreener. exchange_ids: Comma-separated CCXT exchange IDs.

Returns: Object with coin_id, vs_currency, prices (per-source array with source, ok, and either price or error), max, min, spread_bps, n_ok, n_error. Unknown coin_ids still return per-source error envelopes pointing at get_exchange_ticker.

Compute technical indicators on OHLCV candles you have already fetched.

USE THIS WHEN: you have OHLCV data (from get_exchange_ohlcv for one specific exchange, or get_aggregated_ohlc for a CoinGecko cross-venue aggregate) and you want RSI / MACD / Bollinger / EMA / SMA / ATR / ADX / Stochastic / OBV without writing the math yourself.

THIS TOOL DOES NOT FETCH DATA. The caller must provide candles. If you need candles first, call get_exchange_ohlcv (CCXT, per-venue, supports 1m candles) or get_aggregated_ohlc (CoinGecko, market-aggregate, daily/ hourly).

THIS TOOL RETURNS OBSERVATIONS, NOT TRADING ADVICE. The signal_summary field describes what the indicators currently show (e.g. "RSI 72 — overbought"). It never recommends buying or selling.

Input format: ohlcv: list of rows. Either 6 columns: [timestamp_ms, open, high, low, close, volume] (CCXT) 5 columns: [timestamp_ms, open, high, low, close] (CoinGecko aggregated_ohlc — no volume) Rows must be ordered oldest -> newest. With 5-column input, OBV is unavailable (returned as null with a note).

Args: ohlcv: Candles, oldest first. 5 or 6 columns per row. indicators: Which indicators to compute. Any subset of ["rsi", "macd", "bollinger", "ema", "sma", "atr", "adx", "stochastic", "obv"]. Default omits adx/stochastic/obv to keep output compact; pass them explicitly to opt in. rsi_period: Lookback for Wilder's RSI. Default 14. macd_fast / macd_slow / macd_signal: MACD EMA periods. Defaults 12/26/9. bb_period / bb_stddev: Bollinger Bands lookback and stddev multiplier. Defaults 20 and 2.0. ema_periods: List of EMA lookbacks to compute. Default [12,26,50,200]. sma_periods: List of SMA lookbacks to compute. Default [20,50,200]. atr_period: Wilder ATR lookback. Default 14. stoch_k_period / stoch_d_period: Stochastic %K and %D periods. Defaults 14 and 3. adx_period: Wilder ADX lookback. Default 14. include_series: If True, return the full per-bar series for every indicator (suitable for charting). If False (default), return only the latest value per indicator — much smaller payload. When True and the input has more than MAX_SERIES_RETURN (1000) rows, each returned series is truncated to the last MAX_SERIES_RETURN entries and truncated_series_to is set on the response so the caller can tell.

Bounds: The input is rejected with an error if it has more than MAX_OHLCV_ROWS (5000) rows — pass the most recent N bars or split into chunks. This guards against pathological / prompt-injected inputs whose ADX/Stochastic passes and JSON serialization would otherwise dominate runtime and memory.

Returns: Dict with one key per requested indicator plus: - meta: { bar_count, has_volume, last_timestamp_ms, last_close } - signal_summary: human-readable interpretation per indicator (observations only — overbought / oversold / trend direction etc.) Per-indicator shape: - rsi: { latest, period, series? } - macd: { latest: { macd, signal, histogram }, params, series? } - bollinger: { latest: { upper, middle, lower }, period, stddev, percent_b, bandwidth, series? } - ema: { latest: { "12": ..., "26": ... }, periods, series? } - sma: { latest: { "20": ..., "50": ... }, periods, series? } - atr: { latest, period, series? } - adx: { latest: { adx, plus_di, minus_di }, period, series? } - stochastic: { latest: { k, d }, k_period, d_period, series? } - obv: { latest, series? } # null + note if no volume column

Search DEX pairs across all chains by token name, symbol, or address.

Use this when the token is too new or small for CoinGecko (get_price returns empty), or when the user wants DEX-side / on-chain prices specifically. Returns the most-liquid matching pairs first so the top result is usually the "real" market for the token.

Args: query: Free-text query — token name ("BasedPepe"), ticker ("PEPE"), or contract address. DexScreener returns up to 30 pairs per call. limit: Max pairs to return after sorting by USD liquidity desc (1..30).

Returns: Array of trimmed pair objects with fields: chainId, dexId, pairAddress, url, baseToken{address,name,symbol}, quoteToken{...}, priceNative, priceUsd, liquidityUsd, volumeH24, priceChangeH24, fdv, marketCap, pairCreatedAt. On API failure returns {"error": "..."} from the HTTP layer.

Get aggregated OHLC (open/high/low/close) candlestick data across all exchanges.

Use for technical-analysis-style candlestick views of "the market" rather than a single venue. For per-exchange high-granularity candles (1m, 5m, etc.), use get_exchange_ohlcv (CCXT) instead.

Candle width auto-selected by CoinGecko based on days:

• days = 1 -> 30-minute candles

• days in {7, 14, 30} -> 4-hour candles

• days in {90, 180, 365, max} -> daily candles

Args: coin_id: CoinGecko coin ID. vs_currency: Quote currency (e.g. "usd"). days: Window in days. One of "1","7","14","30","90","180","365","max".

Returns: Array of [unix_ms, open, high, low, close] tuples.

Note: coin_id is validated against ^[a-z0-9][a-z0-9._-]{0,127}$.

Get historical total DeFi TVL for one chain, or for all of DeFi combined.

Use to chart "Ethereum TVL over the last year" or "how has total DeFi TVL evolved?" Complements CoinGecko's get_global_defi, which only gives a single current number with no history.

Args: chain: Chain name from list_chains_tvl (e.g. "Ethereum", "Solana", "Base", "Arbitrum"). Empty string returns total TVL across ALL of DeFi (i.e. all chains combined). days: Number of trailing daily points to return (1..3650). Default 90.

Returns: Array of {date, tvl} points, where date is a unix epoch seconds timestamp at UTC midnight and tvl is the chain's total DeFi TVL in USD on that day.

Note: chain (when non-empty) is validated against ^[a-z0-9][a-z0-9._-]{0,127}$.

Get rich metadata for a single coin: description, links, scores, market data, dev/community stats.

Use this when the user wants to learn about a coin (what is it, who built it, links to docs/source/socials), or when you need scores like CoinGecko rank, sentiment up/down vote percentages, or developer activity.

For just the price, use get_price (much cheaper).

Args: coin_id: CoinGecko coin ID, e.g. "bitcoin". localization: Include localized names/descriptions for many languages. Usually false to keep responses small. tickers: Include a tickers array (large). Use get_coin_tickers instead when you specifically want exchange tickers. market_data: Include current price, market cap, 24h/7d/30d/1y change, ATH/ATL, supply, etc. Recommended. community_data: Twitter/Reddit/Telegram follower counts and growth. developer_data: GitHub stars, forks, commit counts, PR activity.

Returns: A coin object with fields like id, symbol, name, description, links, image, market_cap_rank, market_data, community_data, developer_data, categories, genesis_date, etc.

Note: coin_id is validated against ^[a-z0-9][a-z0-9._-]{0,127}$.

List exchange tickers (trading pairs) for a single coin across many venues.

Use to answer "where can I buy/sell X?" or "which exchanges have the best liquidity for X?" Each ticker includes price, volume, bid-ask spread, converted last/volume in BTC/ETH/USD, and CoinGecko's trust score.

Args: coin_id: CoinGecko coin ID. exchange_ids: Optional comma-separated exchange IDs to filter by (CoinGecko exchange IDs, see list_exchanges_directory). page: Pagination page (each page is up to 100 tickers). order: Sort order. Default ranks by liquidity trust score.

Returns: Object with name (coin name) and tickers (array of ticker objects).

Note: coin_id is validated against ^[a-z0-9][a-z0-9._-]{0,127}$.

Get public companies' BTC or ETH treasury holdings.

Useful for "which companies own BTC?" or "what's MicroStrategy's stack?" style questions, and for tracking institutional adoption.

Args: coin_id: Either "bitcoin" or "ethereum" — those are the only assets CoinGecko tracks for public-treasury data.

Returns: Object with total_holdings, total_value_usd, market_cap_dominance, and companies — an array of { name, symbol, country, total_holdings, total_entry_value_usd, total_current_value_usd, percentage_of_total_supply }.

Fetch L2 order books from many exchanges in parallel and merge into one virtual book.

Answers "where is the best bid/ask across the whole market?" — more useful than per-venue get_orderbook for execution analysis. The merged book does NOT aggregate by price level; each level retains its source exchange so the LLM can attribute liquidity per venue.

Args: symbol: CCXT unified symbol (e.g. "BTC/USDT"). Some exchanges may reject the symbol (BadSymbol) — those go into exchanges_error. exchange_ids: Comma-separated CCXT exchange IDs (capped at 10). depth: Levels per side per exchange, clamped to [1, 50].

Returns: Object with: - symbol, depth_per_exchange - exchanges_ok: list of exchange ids that responded - exchanges_error: list of {exchange, error} for failures - bids: merged bids sorted by price desc, capped at depth*len(ok), each entry {price, amount, exchange} - asks: merged asks sorted by price asc, capped likewise - best_bid, best_ask: top of merged book with attribution - spread_bps: (best_ask - best_bid) / mid * 10000

Get full detail for a single DEX pair on a specific chain.

Use after dex_search / get_dex_token_pairs when the user wants the full picture for one specific market — including buy/sell tx counts at multiple windows (m5/h1/h6/h24), price changes, native-token price, base/quote liquidity sides, and links/socials when present.

Args: chain_id: Chain identifier as DexScreener returns it. Typical values: "ethereum", "bsc", "solana", "polygon", "arbitrum", "base", "optimism", "avalanche", "pulsechain", "fantom", "cronos", ... pair_address: The pair's contract address (case-insensitive on EVM).

Returns: The first matching pair object as returned by DexScreener (with the full txns, info, volume, priceChange, liquidity sub-objects), or {"error": "pair not found"} if no pair matches, or the structured HTTP error dict on transport failure.

Note: chain_id is validated against ^[a-z0-9-]{1,32}$ and pair_address against EVM hex / Solana base58.

Get all DEX pairs trading a given token contract, across every chain.

Use this when you have an EVM/Solana contract address and want to see every venue where it's traded — useful for "which DEX has the deepest liquidity for this token?" and for finding the canonical pair on a specific chain. Pairs are returned sorted by USD liquidity desc.

Prefer this over dex_search when you already have the contract address; prefer get_price (CoinGecko) when the token is large and listed on CEXes.

Args: token_address: Token contract address. EVM addresses like "0xA0b86991c6218b36c1d19D4a2e9Eb0cE3606eB48" (USDC on Ethereum) and Solana mint addresses both work. limit: Max pairs to return (1..30 typical). chain: Optional chain filter — only return pairs on this chain. Common values: "ethereum", "bsc", "solana", "polygon", "arbitrum", "base", "optimism", "avalanche". Leave empty for all chains.

Returns: Array of trimmed pair objects (same shape as dex_search). On API failure returns {"error": "..."}.

Note: token_address is validated against EVM hex or Solana base58 (rejects anything containing /, ?, #, .., %, or whitespace).

Get detailed info on a single exchange (CoinGecko directory): description, links, volume, top tickers.

Args: exchange_id: CoinGecko exchange ID (e.g. "binance", "gdax", "kraken"). See list_exchanges_directory to discover IDs. NOTE: CoinGecko exchange IDs sometimes differ from CCXT IDs (e.g. CoinGecko uses "gdax" for Coinbase Pro).

Returns: Exchange object with name, year_established, country, description, url, image, trust score, volume metrics, and a tickers array.

Note: exchange_id is validated against ^[a-z0-9][a-z0-9._-]{0,127}$.

List all trading pairs (markets) supported by a specific exchange.

Use to discover what symbols an exchange trades (e.g. "does Kraken list SOL/USDC?"), or to enumerate available perpetual contracts.

Args: exchange_id: CCXT exchange ID (lowercase). See list_supported_exchanges. active_only: If true, exclude delisted/inactive markets.

Returns: Object with exchange, count, and markets — each market has symbol, base, quote, settle, type (spot/swap/future/option), linear, inverse, contract, active.

Get OHLCV candlestick data from a specific exchange (high-granularity, including 1-minute candles).

Prefer this over get_aggregated_ohlc when:

• the user asks about a specific venue, OR

• they need sub-hour candles (1m/5m/15m), OR

• they need exact volume on one exchange.

Args: exchange_id: CCXT exchange ID, e.g. "binance". symbol: Unified symbol, e.g. "BTC/USDT". timeframe: Candle width. Not every exchange supports every timeframe; common safe choices: "1m","5m","15m","1h","4h","1d". limit: Number of candles. Most exchanges cap at ~500-1500 per call. since_ms: Optional unix-millis lower bound. Most recent candles when null.

Returns: Array of [timestamp_ms, open, high, low, close, volume] tuples, oldest first.

Get a real-time ticker (bid/ask/last/24h stats) for one symbol on one exchange.

Use when the user asks about price on a specific venue ("BTC on Coinbase", "ETH on Binance") or wants tight bid-ask spread info.

Args: exchange_id: CCXT exchange ID, e.g. "binance". symbol: CCXT unified symbol, e.g. "BTC/USDT", "ETH/USD", "BTC/USDT:USDT" for a linear perp on Binance.

Returns: Ticker object with symbol, timestamp, datetime, bid, ask, last, high, low, open, close, vwap, baseVolume, quoteVolume, percentage, change.

Get the Crypto Fear & Greed Index (0=extreme fear, 100=extreme greed).

A widely-quoted contrarian sentiment indicator that combines volatility, momentum, social media, surveys, BTC dominance and trend volume into a single 0-100 score. Useful for "are people fearful or greedy right now?" questions.

Args: limit: How many days of history to return (default 1 = today only, 0 = all history).

Returns: Object with name, data (array of { value, value_classification, timestamp, time_until_update }), and metadata. Higher = greedier.

Get the current funding rate for a perpetual-futures contract on a specific exchange.

Funding rate is a periodic payment between longs and shorts that anchors perp prices to spot. Positive funding means longs pay shorts (crowd is long); negative funding means shorts pay longs.

Args: exchange_id: CCXT exchange ID that supports perps, e.g. "binance", "okx", "bybit", "bitmex". symbol: Perp symbol with settle suffix, e.g. "BTC/USDT:USDT" for the Binance USDT-margined linear perp, "BTC/USD:BTC" for an inverse perp.

Returns: Funding info with symbol, markPrice, indexPrice, fundingRate, fundingTimestamp, nextFundingRate, nextFundingTimestamp, interestRate. Exact fields vary slightly by exchange.

Get historical funding-rate time-series for a perpetual contract on one exchange.

Use this for trend analysis on funding — "is funding turning positive?", "how long has BTC funding been negative?", spotting funding cycles, or feeding a series into a quant signal. For the single most-recent snapshot use get_funding_rate. For a cross-exchange comparison of the current rate use compare_funding_rates.

The funding interval varies by venue: Binance and OKX charge funding every 8 hours, Bybit every 1 hour for some perps, BitMEX every 8 hours, etc. Returned timestamps reflect each charge. Don't assume a uniform cadence when comparing series across exchanges.

Args: exchange_id: CCXT exchange ID supporting perps, e.g. "binance", "okx", "bybit", "bitmex", "bitget", "bingx", "gate", "mexc", "kucoinfutures", "hyperliquid". symbol: Perp symbol with settle suffix. Linear (USDT-margined): "BTC/USDT:USDT". Inverse (coin-margined): "BTC/USD:BTC". since_ms: Optional unix-millis lower bound. If null, the exchange returns its default window (typically the most-recent N rows). limit: Max number of rows to return; clamped to [1, 1000].

Returns: Array of rows oldest-first, each with timestamp (unix millis), datetime (ISO 8601), symbol, fundingRate (e.g. 0.0001 = 1 bp), and exchange-specific extras under info. On unsupported exchanges returns {"error": "..."}.

Get global DeFi market stats: total DeFi market cap, DeFi-to-Eth ratio, top DeFi coin by share.

For protocol-level TVL or chain-level breakdowns, use get_protocol_tvl or list_chains_tvl (DefiLlama) instead — they're much more granular.

Returns: Object with data containing defi_market_cap, eth_market_cap, defi_to_eth_ratio, trading_volume_24h, defi_dominance, top_coin_name, top_coin_defi_dominance.

Get global cryptocurrency market stats: total market cap, total 24h volume, BTC/ETH dominance.

Use for macro questions like "what's the total crypto market cap?", "is BTC dominance rising?", "how many coins exist?".

Returns: Object with data containing: - active_cryptocurrencies, upcoming_icos, ongoing_icos, ended_icos, markets - total_market_cap (mapping of currency -> amount) - total_volume (mapping of currency -> amount) - market_cap_percentage (per-coin share of total cap, e.g. {"btc": 52.1}) - market_cap_change_percentage_24h_usd - updated_at

Get historical price, market cap and total volume time series for a coin.

Use this to draw line charts or compute returns/volatility over a window. For candlestick (OHLC) data use get_aggregated_ohlc instead.

Granularity is auto-selected by CoinGecko based on days:

• days <= 1 -> ~5-minute datapoints

• days <= 90 -> ~hourly datapoints

• days > 90 -> daily datapoints

Args: coin_id: CoinGecko coin ID (e.g. "bitcoin"). vs_currency: Quote currency (e.g. "usd", "eur", "btc"). days: Window in days. Examples: "1", "7", "14", "30", "90", "180", "365", or "max" for the full history. interval: Force daily granularity by passing "daily". Leave empty for auto.

Returns: Object with three arrays of [unix_ms, value] pairs: - prices - market_caps - total_volumes

Note: coin_id is validated against ^[a-z0-9][a-z0-9._-]{0,127}$.

Get detailed data for a single NFT collection: floor price, market cap, volume, holders, links.

Args: nft_id: CoinGecko NFT collection ID, e.g. "bored-ape-yacht-club", "cryptopunks". Use search or list_nfts to find IDs.

Returns: NFT collection object with floor_price, market_cap, volume_24h, floor_price_in_usd_24h_percentage_change, number_of_unique_addresses, total_supply, links, image, etc.

Note: nft_id is validated against ^[a-z0-9][a-z0-9._-]{0,127}$.

Get the current open interest (OI) for a perpetual contract on one exchange.

Open interest is the total notional/contract count of outstanding positions on a venue. Reading OI alongside price:

• OI rising with price rising -> new longs entering, trend has fuel

• OI rising with price falling -> new shorts loading up

• OI falling with price rising -> short squeeze / covering rally

• OI falling with price falling -> longs capitulating This is descriptive, not advice. For funding-rate context use get_funding_rate (snapshot) or get_funding_rate_history (series).

Falls back to fetch_open_interest_history(timeframe="1h", limit=1) when the exchange exposes only the historical endpoint.

Args: exchange_id: CCXT exchange ID supporting perps. symbol: Perp symbol with settle suffix, e.g. "BTC/USDT:USDT" or "BTC/USD:BTC" for inverse.

Returns: Object with symbol, openInterestAmount (in base units / contracts), openInterestValue (notional in quote), timestamp, datetime, and exchange-specific info. On unsupported exchanges returns {"error": "..."}.

Get a Level-2 order-book snapshot (top bids and asks) from a specific exchange.

Use to assess liquidity, spread, and short-term supply/demand on a venue. Note this is a snapshot, not a stream.

Args: exchange_id: CCXT exchange ID. symbol: CCXT unified symbol, e.g. "BTC/USDT". limit: How many price levels per side (default 20). Some exchanges cap this; CCXT will return as many as the venue allows.

Returns: Object with symbol, timestamp, datetime, bids (array of [price, amount] pairs sorted high to low), asks (low to high), nonce.

Get the current spot price of one or more cryptocurrencies.

Use this for the cheapest, fastest "what is X worth right now?" lookup. For historical prices, use get_market_chart or get_aggregated_ohlc. For real-time prices on a specific venue, use get_exchange_ticker.

Args: coin_ids: Comma-separated CoinGecko coin IDs, e.g. "bitcoin,ethereum,solana". IDs are NOT ticker symbols — call search first if unsure. vs_currencies: Comma-separated target currencies, e.g. "usd,eur,btc". include_market_cap: Include each coin's market cap in the response. include_24hr_vol: Include 24h trading volume. include_24hr_change: Include 24h price change percent. include_last_updated_at: Include unix timestamp of last update.

Returns: Mapping of {coin_id: {currency: price, ...}}. When the optional flags are enabled, additional fields like "_market_cap", "_24h_vol", "_24h_change", and "last_updated_at" appear alongside the price.

Get a single protocol's metadata, current TVL, and recent TVL history.

Use after list_protocols to drill into one protocol — e.g. "what is Aave's TVL on each chain?" or "show me Lido's last 90 days of TVL." The raw DefiLlama response is enormous (multi-year daily series for the overall protocol AND every chain it touches), so this tool trims each history series to the last history_days daily points and drops the per-token breakdown arrays (tokens, tokensInUsd).

Args: slug: Protocol slug from list_protocols (e.g. "aave-v3", "lido", "uniswap-v3"). NOT a CoinGecko coin ID. history_days: Number of trailing daily points to keep for each TVL series. Default 90; use a smaller value for compactness or a larger value (up to ~1500) for long-range analysis.

Returns: Protocol object with id, name, symbol, category, chains, chain, description, url, mcap, currentChainTvls (snapshot per chain), tvl (recent series of {date, totalLiquidityUSD}), and chainTvls (per-chain trimmed series).

Note: slug is validated against ^[a-z0-9][a-z0-9._-]{0,127}$.

Get recent public trades (the tape) for a symbol on a specific exchange.

Use to inspect order flow, identify large prints, or compute very short-term trade-driven metrics.

Args: exchange_id: CCXT exchange ID. symbol: Unified symbol, e.g. "BTC/USDT". limit: Number of recent trades to return (max ~1000 depending on venue).

Returns: Array of trades with id, timestamp, datetime, symbol, side, price, amount, cost, takerOrMaker.

Get DefiLlama oracle spot prices for tokens identified by chain:address.

Use when you have a token's contract address and want a price without going through CoinGecko/DexScreener — DefiLlama aggregates DEX prices across many sources. For human-friendly token discovery (search by symbol/name), use dex_search (DexScreener) or CoinGecko's search.

Args: coins: Comma-separated chain:address identifiers, e.g. "ethereum:0xa0b86991c6218b36c1d19d4a2e9eb0ce3606eb48,bsc:0xbb4cdb9cbd36b01bd1cbaebf2de08d9173bc095c". Chain identifiers follow DefiLlama's naming (e.g. "ethereum", "bsc", "polygon", "arbitrum", "base", "solana"). Solana uses "solana:".

Returns: Object with a coins map keyed by chain:address, where each value has decimals, price, symbol, timestamp, confidence.

Note: Each chain:address segment of coins is validated. Chain matches ^[a-z0-9-]{1,40}$ and address is EVM hex, Solana base58, or DefiLlama token id.

Get the biggest price movers (gainers and losers) over a time window.

NOTE: This endpoint typically requires a CoinGecko API key (Demo tier is fine). Without a key it may return an error — fall back to list_top_coins and sort client-side if so.

Args: vs_currency: Quote currency. duration: Time window for price change. top_coins: Universe to scan — "300", "500", "1000", or "all".

Returns: Object with top_gainers and top_losers arrays of coin objects including usd_24h_change (or whichever duration you chose).

Get currently trending coins, NFT collections and categories on CoinGecko (last 24h searches).

Use to answer "what is the market paying attention to right now?" or to surface narratives. This is search-driven, not volume-driven, so it captures emergent interest before price moves.

Returns: Object with arrays coins, nfts, and categories. Each coin item includes item.id, item.name, item.symbol, item.market_cap_rank, and item.data with price/24h-change snapshot.

Parallel-ping every upstream data source and report status + latency.

DEBUGGING tool. Call only when the user reports something is broken ("X is down", "data looks stale") or when you suspect an upstream is degraded. Do NOT call on every request — issues fresh network calls.

Pings in parallel:

• CoinGecko: /ping

• DefiLlama: /protocols?limit=1

• DexScreener: /latest/dex/search?q=BTC

• Alternative.me: /fng/?limit=1

• CCXT/Binance: fetch_status() (fetch_ticker fallback)

Returns: Object with checked_at (ISO8601 UTC), all_ok (bool — true iff every source succeeded), and sources — array sorted by latency asc. Each entry has source, ok, latency_ms, and detail (short success identifier or <ExceptionType>: <first 100 chars> on failure).

List all coin categories with aggregated market data (market cap, 24h volume, 24h change).

Use to discover narratives ("Layer 1", "DeFi", "Meme", "AI", "RWA", "Liquid Staking", ...) and to find category IDs you can pass to list_top_coins(category=...).

Args: order: Sort order.

Returns: Array of categories with id, name, market_cap, market_cap_change_24h, volume_24h, top_3_coins, updated_at.

List blockchain networks ranked by current total DeFi TVL.

Use for "which chains have the most DeFi activity?", "how does Solana's TVL compare to Ethereum's?", or to discover chain names you can pass to get_chain_tvl_history or list_protocols(chain=...).

Args: limit: Number of chains to return after sorting by TVL desc (1..200).

Returns: Array of chain summaries with name, tvl, tokenSymbol, chainId, gecko_id, cmcId.

List derivatives (futures/perp/options) exchanges ranked by open interest or volume.

Use to compare derivatives venues (Binance Futures, Bybit, OKX, dYdX, etc.) by size.

Args: order: Sort order. per_page: 1..250 per page. page: Page number.

Returns: Array of derivatives exchanges with id, name, open_interest_btc, trade_volume_24h_btc, number_of_perpetual_pairs, number_of_futures_pairs, year_established, country, url.

List DEXes ranked by 24-hour trading volume.

Use for "biggest DEXes by volume", "Uniswap vs PancakeSwap volume", or to see momentum (each entry includes 1d/7d/30d change percentages). Pair with CCXT's get_exchange_ticker for centralized-exchange volumes.

Args: limit: Number of DEXes to return (1..200).

Returns: Object with a summary (totals across all DEXes: total24h, total7d, total30d, change_1d, change_7d, change_1m) and protocols — an array of name, displayName, slug, category, chains, total24h, total7d, total30d, total1y, totalAllTime, change_1d, change_7d, change_1m.

List centralized exchanges from CoinGecko's directory, ranked by trust score / volume.

This is CoinGecko's curated directory with metadata (year established, country, trust scores, 24h BTC-equivalent volume). For exchanges you can actually query in real time via this MCP, see list_supported_exchanges.

Args: per_page: 1..250 exchanges per page. page: Page number.

Returns: Array of exchanges with id, name, year_established, country, url, image, trust_score, trust_score_rank, trade_volume_24h_btc, trade_volume_24h_btc_normalized.

List protocols ranked by fees or revenue (daily or all-time).

Useful for "which protocols make the most money?", "Tron vs Ethereum fees", or "Aave's revenue this month". DefiLlama distinguishes fees (gross paid by users) from revenue (kept by the protocol/token holders), and offers daily and all-time aggregations — pick via data_type.

Args: limit: Number of protocols to return (1..200). data_type: Which metric series to pull. "dailyFees" = fees on a daily basis (typical for "what's hot today?"), "dailyRevenue" = revenue on a daily basis, "totalFees"/"totalRevenue" = cumulative all-time. Note: regardless of data_type, each protocol's per-window fields are still named total24h, total7d, total30d, etc., but they now refer to the chosen metric.

Returns: Object with a summary (totals across all protocols: total24h, total7d, total30d, change_1d, change_7d, change_1m) and protocols — an array of name, displayName, slug, category, chains, total24h, total7d, total30d, total1y, totalAllTime, change_7dover7d, change_30dover30d.

List the latest tokens that have set up a profile on DexScreener.

A profile means the project has filled in description / website / socials on DexScreener — typically a sign of a newly-launched but at least somewhat promoted token. Useful as an early-signal feed for "what's new today?"

NOTE: a profile does not imply legitimacy or liquidity. Cross-check with get_dex_token_pairs(token_address, ...) before quoting prices.

Args: limit: Max tokens to return (the API itself returns up to ~30).

Returns: Array of token-profile objects with chainId, tokenAddress, url, description, icon, and a links array of websites/socials. On API failure returns {"error": "..."}.

List NFT collections sortable by floor price, market cap, or 24h volume.

Args: order: Sort order. per_page: 1..250. page: Page number.

Returns: Array of NFT collections with id, name, symbol, asset_platform_id, contract_address. Use get_nft_collection for details on one.

List DeFi protocols ranked by TVL (or 1d/7d change, or market cap).

Use this for "what are the biggest DeFi protocols?", "which protocols had the largest TVL inflows/outflows today?", or to find a protocol's slug before calling get_protocol_tvl. Far more granular than CoinGecko's get_global_defi, which only returns aggregate DeFi market cap.

Args: limit: Number of protocols to return after sorting (1..500). sort_by: Metric to rank by, descending. "tvl" = current TVL, "change_1d"/"change_7d" = TVL change %, "mcap" = token market cap. chain: Optional chain name filter (e.g. "Ethereum", "Solana", "Base", "Arbitrum"). Matches against each protocol's chains array, case-insensitive. Empty string disables the filter.

Returns: Array of protocol summaries with name, slug, symbol, category, chain, chains, tvl, change_1h, change_1d, change_7d, mcap, url.

Note: chain (when non-empty) is validated against ^[a-z0-9][a-z0-9._-]{0,127}$.

List stablecoins ranked by current circulating market cap.

Use for "what are the biggest stablecoins?", "is USDT or USDC bigger?", "where is USDC issued (which chains)?", or to track peg health (price field shows the current oracle price).

Args: limit: Number of stablecoins to return (1..200). include_chain_breakdown: If true, includes a chainCirculating map of supply per chain. If false, drops it to keep the response small (the breakdown for the top stablecoins is verbose).

Returns: Array of stablecoin summaries with id, name, symbol, pegType, pegMechanism, price, circulating, circulatingPrevDay, circulatingPrevWeek, circulatingPrevMonth, chains, and (optionally) chainCirculating. The id field is what get_stablecoin_detail would accept (DefiLlama internal id).

List all exchange IDs that this server can query in real time via CCXT.

Use this when you need to know which exchange_id values are valid for the other CCXT-backed tools (get_exchange_markets, get_exchange_ticker, get_orderbook, get_recent_trades, get_exchange_ohlcv, get_funding_rate).

Returns: Object with count and exchanges (array of lowercase exchange IDs such as "binance", "okx", "coinbase", "kraken", "bybit", "kucoin", "huobi", "bitfinex", "gateio", "mexc").

List currently top-boosted tokens on DexScreener (paid promotion).

"Boosts" are paid promotion slots — projects pay to surface their token. Read these results with skepticism: high boost spend does NOT imply quality, traction or safety. This list is mostly useful as a signal of "what is being actively pushed right now" rather than "what is good".

For an unbiased early-tokens view prefer list_latest_dex_tokens; for actual price/liquidity always confirm via get_dex_token_pairs.

Args: limit: Max tokens to return.

Returns: Array of boosted-token objects with chainId, tokenAddress, url, description, icon, links, and totalAmount (boost spend). On API failure returns {"error": "..."}.

List top coins with market data (price, market cap, volume, change %) — sortable and paginated.

The workhorse for "show me the top N coins" / "top by volume" / "top in DeFi" style questions. Each call returns up to 250 coins; paginate for more.

Args: vs_currency: Quote currency (e.g. "usd"). order: Sort order. Default is descending market cap. per_page: 1..250 coins per page. page: Page number, 1-indexed. category: Optional category ID to filter by (see list_categories), e.g. "decentralized-finance-defi", "layer-1", "meme-token". price_change_percentages: Comma list of windows to include — any of "1h,24h,7d,14d,30d,200d,1y".

Returns: Array of coin objects with id, symbol, name, image, current_price, market_cap, market_cap_rank, total_volume, high_24h, low_24h, price_change_*_in_currency, ath, atl, circulating_supply, etc.

List DeFi yield-bearing pools (lending, staking, LPs) filtered & ranked by APY or TVL.

The DefiLlama yields endpoint returns ~20k pools, so this tool aggressively filters and slices client-side. Use for "best stablecoin yields right now", "highest APY on Aave", "Lido vs Rocket Pool TVL", etc. Pair with list_protocols if you want protocol-level TVL rather than per-pool APY.

Args: min_tvl_usd: Minimum pool TVL in USD. Default 1M filters out tiny pools. Set to 0 to disable. project: Optional project filter, e.g. "aave-v3", "lido", "compound-v3". Matches the project field, case-insensitive. chain: Optional chain filter, e.g. "Ethereum", "Solana", "Arbitrum". Case-insensitive exact match against chain. symbol: Optional symbol/token substring filter, e.g. "USDC", "ETH", "STETH". Case-insensitive substring match against symbol. limit: Number of pools to return after filtering & sorting (1..500). sort_by: Metric to rank by, descending. "apy" = current APY, "tvlUsd" = pool size, "apyMean30d" = 30-day mean APY.

Returns: Array of pool summaries with pool (DefiLlama pool id), chain, project, symbol, tvlUsd, apy, apyBase, apyReward, apyMean30d, apyPct1D, apyPct7D, apyPct30D, stablecoin, ilRisk, exposure, predictions, rewardTokens, underlyingTokens.

Note: project/chain/symbol (when non-empty) are validated. project and chain use ^[a-z0-9][a-z0-9._-]{0,127}$; symbol allows mixed case via ^[A-Za-z0-9._-]{0,63}$.

Universal CoinGecko search — resolves names/symbols to IDs across coins, exchanges, categories and NFTs.

ALWAYS use this first when the user mentions a coin/exchange/NFT by name or ticker symbol and you don't already know the canonical CoinGecko ID. Most other tools require IDs.

Args: query: Free-text query, e.g. "btc", "uniswap", "bored ape".

Returns: Object with arrays: - coins: [{ id, name, symbol, market_cap_rank, ... }] - exchanges: [{ id, name, market_type }] - categories: [{ id, name }] - nfts: [{ id, name, symbol }]