mcp.bitquery.io now lives here: same server, same endpoint, new home in Products. Read the docs →
MCP SERVER · Model Context Protocol · AI agents

Bitquery MCP Server for crypto trading data and investigations in AI agents.

A hosted crypto MCP server. Add https://mcp.bitquery.io to Claude, ChatGPT, Cursor, VS Code or Codex, sign in once with your Bitquery account, and ask on-chain questions in plain English. The agent picks from 256 typed tools for trending tokens, candles, top traders, Polymarket odds, Hyperliquid positions, money-flow tracing, address labels and raw SQL, backed by the same index that serves Bitquery's APIs on 40+ chains.

endpointhttps://mcp.bitquery.io

Cursor and VS Code install with one click. Claude Code copies the command. Claude Desktop, ChatGPT and Codex take the URL in four steps · Talk to sales

MCPStreamable HTTPOAuth 2.1256 tools
Trusted by 40,000+ developers & teams like
Binance logoChainalysis logoTRM Labs logoNansen logo0x logoCoinMarketCap logoCoin Metrics logoBybit logoLukka logo3Commas logoNexo logoTether logo
40+
Chains supported
1PB+
Blockchain data indexed
10B+
API calls / month
99.9%
Production uptime
01
Install in 60 seconds

How do you connect the Bitquery MCP server to Claude, ChatGPT, Cursor or VS Code?

The Bitquery MCP server is a hosted Model Context Protocol server at https://mcp.bitquery.io. It speaks Streamable HTTP and signs you in with OAuth 2.1 against your Bitquery account, so there is no config file to edit and no API key to paste: add the URL as a custom connector or MCP server in your client, approve the login in the browser once, and the tools appear in the conversation.

Cursor and VS Code install with one click. Claude Code has no one-click link yet, so its button copies the command. ChatGPT and Codex follow the steps below. Full guide: docs.bitquery.io/docs/mcp/mcp-server

Cl

Claude Desktop & Web

  1. Open Settings → Connectors.
  2. Click Add custom connector.
  3. Name: Bitquery. URL: https://mcp.bitquery.io.
  4. Click Add. A browser window opens to log in with your Bitquery account.
CC

Claude Code (CLI)

claude mcp add --transport http bitquery https://mcp.bitquery.io

Append --scope user to enable it in every project. On first use run /mcp inside Claude Code (or claude mcp login bitquery) and complete the browser login.

GP

ChatGPT

  1. Open Settings → Connectors (Plus, Pro or Business).
  2. Click Add connector → Custom connector.
  3. URL: https://mcp.bitquery.io.
  4. Sign in with your Bitquery account.
Cu

Cursor

  1. Open Settings → MCP.
  2. Click Add new MCP server.
  3. URL: https://mcp.bitquery.io.

Or skip the steps with the Install in Cursor button above.

Cx

OpenAI Codex (CLI)

codex mcp add bitquery --url https://mcp.bitquery.io

Or add it to ~/.codex/config.toml:

[mcp_servers.bitquery] url = "https://mcp.bitquery.io"

The first call opens a browser for the OAuth login.

VS

VS Code and other MCP clients

Add an MCP server with the URL https://mcp.bitquery.io, or use the Install in VS Code button above. Windsurf and custom agents built on the Anthropic or OpenAI SDKs connect the same way.

A client that only speaks stdio can bridge through mcp-remote:

npx -y mcp-remote https://mcp.bitquery.io/mcp
02
Two datasets, two plans

What can an AI agent do with the Bitquery MCP server?

The catalog splits into trading and market data on one side and crypto investigations on the other. The two MCP plans follow the same line: Trading MCP unlocks the market tools, AI Investigation MCP unlocks everything, labels and tracing included.

Trading MCP · 32 tools

Trading and market data

A trading MCP for AI agents: every DEX trade on 9 chains, indexed to the second, including tokens that never reach CoinGecko or CoinMarketCap.

  • Trending tokens, gainers, losers and the most volatile, per chain
  • OHLCV candles from 1 second to 1 hour, per token or per pair
  • Top, most profitable and quietly accumulating traders, with PnL
  • Token dashboards, asset comparisons and cross-chain arbitrage spreads
  • Raw SQL over 30 days of per-second trades, on paid plans
DEX trades on
EthereumArbitrumBasePolygonOptimismBNB ChainTronSolanaRobinhoodArc
AI Investigation MCP · all 256 tools

Crypto investigations

From one address to the whole story: money flows, counterparty labels, risk verdicts and flow maps, in about a minute.

  • Trace where a wallet sent or received funds, hop by hop
  • Name the counterparties: exchanges, mixers, scams, unknowns
  • AML risk screening with a LOW, MEDIUM or HIGH verdict and the evidence
  • Money-flow diagrams and full address dossiers in one run
  • Millions of address labels in both directions: identify one address, or list every address tagged mixer, sanctioned or scam
Transfer tracing on
EthereumBNB ChainPolygonArbitrumBaseOptimismRobinhoodTronSolanaBitcoinZcashCardanoXRP Ledger
03
Beyond DEX trades

Does it cover Polymarket, Hyperliquid and Solana?

Yes. Three tool families go past token swaps, each with worked examples in the docs. The figures below came from the server on 21 September 2026.

Polymarket prediction markets

14 read-only tools for odds, price history, fills, wallets and holders. That day no market was in both the top five by volume and the top five by wallets, and the week's eighth-largest wallet won three of six bets and was still down about $434,000. Polymarket MCP server, whale tracking, analytics.

Hyperliquid

18 tools for funding, open interest, liquidations, account rankings and positions. After Bitcoin rose 5.4% in 24 hours, 9,265 of the 9,334 BTC liquidations in that window were shorts, and BTC funding went from 10.95% to 78.64% per year in two hours. Hyperliquid MCP server.

Solana and pump.fun

New launches, the first seconds of a launch, first buyers, and transfer tracing between wallets. The top "USDT" in Solana's 24-hour volume ranking that day was a counterfeit with 5 trades between 2 wallets, and the guide shows how to catch it. Solana MCP server.

Polymarket & Hyperliquid Data in ChatGPT and Claude | Bitquery MCP. A 38-second demo of Polymarket, Hyperliquid and DEX questions answered in plain English.

Guides and examples

04
Watch

What does an investigation look like inside Claude?

Most crypto investigation tools cost thousands of dollars a month. The Bitquery MCP server does the same job inside Claude or ChatGPT: paste a wallet address, ask your question in plain English, and get the answer back with the evidence.

Introducing Bitquery MCP for Crypto Investigations. We point it at a single Tron wallet. It pulls the full transaction history, labels the counterparties, tells known exchanges apart from unknown wallets, traces where the money moved and writes a verdict: what the wallet is doing, how risky it looks, and the leads worth chasing next. The money-flow map at the end comes out of the same run.

How to connect Bitquery MCP. The setup walkthrough: adding the server as a connector, the OAuth sign-in with a Bitquery account, and the first on-chain questions.

05
Ask in plain English

What can you ask the Bitquery MCP server?

Each question below maps to one tool or one guided prompt. The agent picks it; you never see the call unless you ask to.

Ask it like an investigator

"Trace where the funds from this wallet went and stop at the first exchange."

runs the money_flow prompt

"Screen this address for mixer, sanctioned or scam exposure and give a risk verdict."

runs the risk_screen prompt

"Build a full dossier on this address: activity, transfers, labels, likely same-owner wallets."

runs the address_report prompt

"Which entity is behind this label, what addresses does it control, and who does it transact with?"

runs the entity_report prompt

"Is this wallet an exchange deposit address? A mixer? A known scam?"

uses address_labels: entity, category, CEX deposit or hot wallet, mixer, gambling, scam, darknet and token-clone labels

"Expand this Bitcoin address into the addresses that likely belong to the same owner."

uses btc_related_addresses: common-input-ownership clustering

Ask it like a trader

"What are the trending Solana tokens in the last 24 hours by trade volume?"

uses trending_tokens

"Show me OHLCV candles for WETH/USDC on Uniswap v3 for the past 7 days."

uses pair_ohlcv

"Who are the most profitable traders of this token this month, and what are they holding now?"

combines profitable_traders_by_token and trader_positions

"Build a profile of wallet 0xabc…: total PnL, win rate, top tokens, recent activity."

uses trader_profile and trader_activity

"Find every token whose symbol contains 'AI' on Base, sorted by 24h volume."

uses find_tokens

"What are the odds on this Polymarket market, and how did they move this week?"

combines polymarket_market_odds and polymarket_market_price_history

"Which Hyperliquid markets had the most liquidations today, and on which side?"

uses hyperliquid_liquidations

"Which wallets are quietly accumulating this token right now, buys far above sells?"

uses accumulating_traders_by_token

"Decode this transaction: every swap inside it, with tokens, amounts, price and pool."

uses tx_trades

"Show the top holders of this token contract on Ethereum."

uses eth_token_holders (also Polygon, Arbitrum, Base, Optimism)

"Run raw SQL over the last 30 days of per-second DEX trades and the address-label directory."

uses execute_sql (paid plans)
06
Guided prompts

Which guided prompts does the server ship?

Guided prompts are one-shot workflows (MCP prompts/) that chain several tools and return a chart, a table or a diagram. Pick them from your client's prompt or command menu.

Investigations

money_flow

Trace funds from an address or transaction across hops, label the counterparties, stop at exchanges or mixers. Returns a money-flow diagram and an edge table.

risk_screen

AML screen of an address or transaction for mixer, sanctioned, scam and darknet exposure. Returns a LOW, MEDIUM or HIGH verdict, the evidence and an exposure diagram.

address_report

Full dossier for one address: activity stats, recent transfers, labels and, on Bitcoin, likely same-owner addresses.

entity_report

Resolve a labeled entity to its addresses, then their activity and connections.

Trading

token_chart

Resolve a token and render a candlestick or line price chart with a data table.

token_dashboard

360° view of one token: price chart, supply and FDV, top and smart-money traders, scam check.

trader_dashboard

One wallet: profile stats, holdings and PnL positions, activity time series.

compare_assets

Overlay 2 to 5 assets on one normalized %-change chart with a stats table.

token_scout

Trending, gainers, momentum and accumulation, ranked into a pivot of promising tokens.

market_movers

Chain leaderboard: top gainers and most traded, as a sortable table and bar chart.

arbitrage_scan

One asset across chains: a price-spread % table and bar chart.

07
256 tools under the hood

Which tools does the Bitquery MCP server expose?

Every prompt and example above is built from individual tools the agent can also call directly. The catalog groups into eight families. The full list, parameter shapes and supported networks come from the server itself when your client calls tools/list, so the agent always sees the current catalog. Family list checked 21 September 2026.

families
tools/list · one turn, three toolsmcp.bitquery.io
# "Who are the most profitable traders of PEPE this month,
#  and what are they holding now?"

find_tokens                  → "PEPE" → contract + chain
profitable_traders_by_token  → ranked by realized + unrealized PnL
trader_positions             → current holdings, per wallet
ToolFamilyWhat it does
find_tokensDiscoveryname or ticker (wildcards) → contract + chain
chain_capabilitiesDiscoverythe per-chain tool matrix
pair_ohlcvPricescandles from 1 second to 1 hour, USD or quote
trending_tokensTradersvolume, gainers, losers, most volatile per chain
trader_profileTradersPnL, win rate, top tokens, recent activity
polymarket_top_marketsPolymarketmarkets ranked by volume, orders, wallets or fees
hyperliquid_liquidationsHyperliquidforced closes, one by one or per market
eth_trace_dominant_pathTracingfollows the largest edge, up to 5 hops
btc_related_addressesTracingcommon-input-ownership clustering
address_labelsLabelsentity, CEX deposit, mixer, scam, darknet, clone
execute_sqlRaw SQLread-only ClickHouse, 30 days of trades (paid)

Discovery

find_tokens resolves a name or ticker (wildcards allowed) to contract and chain; find_currencies, token_chains and token_dex_venues map an asset across chains and DEXs; chain_capabilities returns the per-chain tool matrix.

Prices & charts

token_price and token_ohlcv, pair_ohlcv in USD or the quote token, cross-chain currency_* aggregates for majors, plus supply, market cap and pool_recent_trades. Candles from 1 second to 1 hour.

Traders

trending_tokens by volume, gainers, losers or volatility; top_traders by token, pair or network; profitable_ and accumulating_traders; per-wallet trader_profile, positions and activity.

Polymarket

14 read-only tools: polymarket_top_markets and polymarket_find_markets, polymarket_market_odds and polymarket_market_price_history from 1 second to 1 day, polymarket_market_trades, wallet tools with positions and profit, polymarket_outcome_holders, settlements and disputes.

Hyperliquid

18 tools for perpetuals, spot and third-party markets: hyperliquid_markets, price and candles, hyperliquid_funding_rates and hyperliquid_open_interest, hyperliquid_liquidations, TWAP orders, hyperliquid_top_traders, positions and fills.

Tracing

A per-chain family for 13 chains (eth_, bsc_, matic_, arbitrum_, base_, optimism_, robinhood_, tron_, solana_, btc_, zec_, ada_, xrp_): *_address_flow_summary, *_flow_edges, *_trace_next_hop, *_trace_dominant_path, *_transfers_in and _out, *_tx_transfers, *_token_holders, plus contract *_find_events and *_find_calls.

Labels

address_labels for one address, labels_for_addresses for a batch, addresses_by_label for every address tagged X, find_label_values to discover what exists: entities, CEX deposit and hot wallets, mixers, gambling, scams, darknet, token clones.

Raw SQL

execute_sql runs read-only ClickHouse over 30 days of per-second DEX trades plus the label directory; *_transfers_raw_sql does the same on each chain's transfer data. Paid plans.

*_address_flow_summary is the one-call triage of any hop: transfer counts, top senders and receivers, and a computed role (consolidator, distributor, relay, or hub at exchange scale, where tracing should stop). *_trace_dominant_path follows the largest edge hop by hop, up to 5 hops in one call. Per-chain coverage varies slightly: Solana adds solana_find_instructions, Bitcoin adds btc_related_addresses and btc_tx_flow with change outputs flagged; chain_capabilities is the authoritative matrix.

08
MCP or GraphQL?

Why use the MCP server instead of the Bitquery GraphQL API?

The dataset is the same Bitquery index behind the GraphQL API, Kafka streams and TradingView feeds. What changes is the interface: the agent picks a typed tool and supplies the arguments, and you stay in the conversation. When you need a query you control line by line, or a stream, the GraphQL API and Kafka are still there on the same account.

What your agent needs
Do it yourself
Bitquery MCP server
Ask on-chain questions
Hand-write GraphQL per question
Plain English; the agent picks the tool
Chain several lookups
Glue the queries together yourself
Typed tool definitions the model chains in one turn
Connect to your LLM client
Build a custom integration
Standard MCP over Streamable HTTP
Long answers
Requests time out
Streaming transport for big histories and multi-step chains
Multi-chain in one place
A pipeline per chain
One endpoint, one catalog
Secure access
API keys pasted into prompts
OAuth 2.1 with your Bitquery account
Billing
A second vendor to manage
Same account: an MCP plan, or credit top-ups on a paid API plan

What teams say about our data

"We did a thorough search of the market for the best onchain data. Bitquery came out on top — and now powers all live prices across Nansen. We don't think of them as a vendor. They're a partner."

A
Alexander Karsten
Nansen
256
MCP tools in eight families
13
Chains with hop-by-hop transfer tracing
1s
Finest candle interval on DEX trades

Bitquery does the hard work of parsing blockchain transaction data into a usable form so that we don't have to. We use their interface to diagnose issues with complex transactions and their analytics as a starting point for our own.

0x Protocol logo
Alex Knaggs
0x Protocol

They proved they had the technology to deliver sophisticated data solutions. We extended our support through the Binance X fellowship — building an open-source library of visualization widgets on their blockchain data.

Director, Binance X logo
Flora Sun
Director, Binance X

The complex raw data is available at different levels of detail and from different viewpoints — whether we need simple aggregated transfers or parameters for failed contract calls. The support is responsive, friendly and quick.

Backend Developer, Blockpit logo
Jan Dreske
Backend Developer, Blockpit

Partnering with Bitquery has been highly cost-effective — leveraging their established infrastructure rather than building our own let us rapidly expand our blockchain support and reach a much broader segment of on-chain users.

Co-Founder, Syla logo
Nick Christie
Co-Founder, Syla

Bitquery's products are very intuitive and easy to use. We currently use their products to obtain DEX-related trading and liquidity information, which saves us the manpower and tedious technical details required to develop our own system. Their excellent technical team deserves special praise; they provide near-24/7 support and resolve issues quickly. I greatly appreciate their products and work ethic.

Ourbit logo
Data Team
Ourbit

Bitquery provides the infrastructure we rely on every day. Fast, reliable, and comprehensive across the chains that matter to our business.

Webacy
Webacy
webacy.com
09
Pricing

What does the Bitquery MCP server cost?

MCP is sold on its own, in credits. Every 7-day trial includes 100 free credits, and any paid API plan can top up credits without switching plans.

Yearly takes about 20% off
Free trial
100 credits
$0
7-day trial, no card

Included with the 7-day trial, no card required. Enough to connect a client and run your first prompts.

Plan · Trading MCP
25k credits / mo
$19 /mo
$15/mo billed yearly
32 tools

Trading data only: prices, OHLCV, DEX trades, top traders and trader activity, for AI agents.

Plan · AI Investigation MCP
200k credits / mo
$149 /mo
$119/mo billed yearly
All 256 tools

All data + address labels: money-flow tracing, address profiles and onchain SQL for AI agents.

Top-up · any paid API plan
300k or 1M credits / mo
$69 / $199 /mo
$55.20 / $159.20 billed yearly

Add MCP credits to a paid API plan without changing it. Recurring, cancel any month.

Get MCP access →Compare all plansPrices in USD. Yearly billing takes about 20% off.
FAQ

Bitquery MCP server questions, answered.

What is the Bitquery MCP server?
A hosted Model Context Protocol server at https://mcp.bitquery.io that exposes Bitquery's on-chain data as MCP tools. An AI client such as Claude, ChatGPT, Cursor, VS Code or Codex calls those tools during a conversation, so you ask in plain English and the agent returns real rows: DEX trades, prices, top traders, Polymarket odds, Hyperliquid positions, transfers, money flow and address labels.
How do I install the Bitquery MCP server?
Add https://mcp.bitquery.io as a custom connector or MCP server in your client and sign in once with your Bitquery account. In Claude Code run: claude mcp add --transport http bitquery https://mcp.bitquery.io. In OpenAI Codex run: codex mcp add bitquery --url https://mcp.bitquery.io. Cursor and VS Code have one-click install buttons on this page; ChatGPT takes the URL under Settings, Connectors, Custom connector.
Which AI clients can connect to it?
Any client that supports remote MCP servers over Streamable HTTP: Claude Desktop and Claude.ai (Plus, Pro or Max), Claude Code, ChatGPT (Plus, Pro or Business with custom connectors enabled), Cursor, VS Code with MCP enabled, OpenAI Codex, Windsurf, and custom agents built with the Anthropic or OpenAI SDKs. A client that only speaks stdio can bridge through mcp-remote.
Can I use it for crypto investigations and AML compliance?
Yes. The server ships guided investigation prompts: money_flow traces funds across hops and stops at exchanges or mixers, risk_screen returns a LOW, MEDIUM or HIGH AML verdict with evidence, address_report builds a full dossier for one address, and entity_report resolves a labeled entity to its addresses and activity. Transfer tracing covers Ethereum, BNB Chain, Polygon, Arbitrum, Base, Optimism, Robinhood, Tron, Solana, Bitcoin, Zcash, Cardano, XRP Ledger, backed by millions of address labels.
Which blockchains and datasets are available?
DEX trading tools cover Ethereum, Arbitrum, Base, Polygon, Optimism, BNB Chain, Tron, Solana, Robinhood, with candles down to 1-second granularity, and Arc for market data. Transfer-tracing tools cover Ethereum, BNB Chain, Polygon, Arbitrum, Base, Optimism, Robinhood, Tron, Solana, Bitcoin, Zcash, Cardano, XRP Ledger. Separate tool families cover Polymarket prediction markets and the Hyperliquid exchange. Tools span token prices, OHLCV candles, trending tokens, top, profitable and accumulating traders, trader profiles and positions, DEX pairs, transfers, the address-label directory, and raw SQL on paid plans.
Is there an MCP server for Polymarket?
Yes. The Bitquery MCP server includes 14 read-only Polymarket tools: markets ranked by volume or wallets, current odds, price history from 1 second to 1 day, single fills, top wallets, positions with profit and loss, holders, settlements and disputed resolutions. It cannot place orders. Worked examples are in the docs at https://docs.bitquery.io/docs/mcp/polymarket/.
Is there an MCP server for Hyperliquid?
Yes. 18 tools cover Hyperliquid perpetuals, spot pairs and third-party markets: prices and candles, funding and open interest, liquidations, TWAP orders, account rankings, positions and fills. It is read-only. Worked examples are at https://docs.bitquery.io/docs/mcp/hyperliquid/.
Does it cover Solana and pump.fun?
Yes. The market tools cover Solana DEX trades, new pump.fun launches, the first seconds of a launch and its first buyers, and a family of solana_ tools traces transfers between wallets. Worked examples are at https://docs.bitquery.io/docs/mcp/solana/.
How much does the Bitquery MCP server cost?
MCP has its own self-service plans: Trading MCP is $19/mo ($15/mo billed yearly) with 25k MCP credits a month and 32 tools; AI Investigation MCP is $149/mo ($119/mo billed yearly) with 200k MCP credits a month and all 256 tools. Every trial includes 100 free credits, and any paid API plan can top up 300k credits for $69/mo or 1M credits for $199/mo.
How does authentication work?
OAuth 2.1. The AI client opens a browser to your Bitquery account, you approve access, and tokens refresh automatically for about 30 days. Access is tied to your Bitquery account, so revoking it there disables the connection at once. No API key is pasted into a prompt or a config file.
How is it different from the Bitquery GraphQL API?
The MCP server is built for AI agents. Instead of writing a GraphQL query, the agent picks an MCP tool, for example token_ohlcv or top_traders, and the server runs the underlying query for you. The data source is the same; the interface is conversational. Billing runs through the same Bitquery account, on an MCP plan or as credit top-ups to a paid API plan.
How is it different from the CoinGecko or CoinMarketCap MCP servers?
Those servers answer market-level questions from their listings: prices, market caps, rankings and trends, behind an API key. The Bitquery server reads on-chain activity directly, so the agent gets individual trades, wallet balances, holders, transfers and fund flows for any token, listed or not, and signs in with OAuth instead of a key.
How fresh is the data?
Near real time. The dataset updates as blocks are processed across all supported chains, so your agent reads both live and historical on-chain data from the same tools.

Give your agent on-chain data today.

Add https://mcp.bitquery.io to Claude, ChatGPT, Cursor, VS Code or Codex and ask in plain English. Nothing to install on a server and no GraphQL to write.

Hosted · OAuth 2.1 · Streamable HTTP · plans from $19/mo