Bitquery MCP Server — Trading Data & Crypto Investigations for AI Agents

Q: What is the Bitquery MCP Server?

It is a hosted Model Context Protocol (MCP) server at https://mcp.bitquery.io that exposes Bitquery's on-chain blockchain data as MCP tools. AI agents like Claude, ChatGPT, Cursor, and VS Code can call these tools directly during a conversation, without you having to write GraphQL queries by hand.

Q: 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/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, Polygon, Arbitrum, Base, Optimism, Tron, Bitcoin and Solana, backed by millions of address labels.

Q: How does authentication work?

Authentication uses OAuth 2.1. The AI client opens a browser to your Bitquery account, you approve access, and tokens refresh automatically for ~30 days. Access is tied to your Bitquery account, so revoking it there immediately disables the connection.

📈Trading & Market Data

Every DEX trade on 8 chains, indexed down to the second — including tokens that never reach CoinGecko.

Trending tokens, gainers, losers and most-volatile per chain
OHLCV candles from 1-second to 1-hour granularity, per token or pair
Top, most profitable and quietly accumulating traders, with PnL
Token dashboards, asset comparisons, cross-chain arbitrage spreads

See trading examples ↓

🔍Crypto Investigations

From one address to the whole story — money flows, labels, risk verdicts, flow maps.

Trace where a wallet sent or received funds, hop by hop
Identify counterparties: exchanges, mixers, scams, unknowns
AML risk screening with LOW / MEDIUM / HIGH verdicts + evidence
Money-flow diagrams and full address dossiers in one run

See a full investigation ↓

Crypto Investigations

Follow the money inside your AI client

Most crypto investigation tools cost thousands of dollars a month. Bitquery MCP 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.

In the video above we point it at a single Tron wallet and let it work: it pulls the full transaction history, labels the counterparties, tells known exchanges apart from unknown wallets, traces where the money actually moved, and writes a clear verdict — what the wallet is doing, how risky it looks, and the leads worth chasing next. The money-flow map at the end is generated in the same run.

40+blockchains indexed

Millionsof address labels

8 chainswith transfer tracing

~1 minfrom address to story

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 / 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 / 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

Guided investigation prompts

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

money_flow

Trace funds from an address or transaction across hops, label counterparties, stop at exchanges / mixers → money-flow diagram + edge table.

risk_screen

AML / compliance screen of an address or transaction (mixer / sanctioned / scam / darknet) → LOW / MEDIUM / HIGH verdict + evidence + exposure diagram.

address_report

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

entity_report

Resolve a labeled entity → its addresses → activity and connections.

The tracing toolkit underneath

The prompts are built from per-chain tracing tools your agent can also call directly — the same family exists for each traced chain (eth_, base_, arbitrum_, matic_, optimism_, tron_, solana_, btc_):

*_address_flow_summary — one-call triage of any hop: transfer counts, top senders & receivers, and a computed role — consolidator, distributor, relay, or hub (exchange-scale; stop tracing there).
*_flow_edges — money-flow graph edges (source → target, amount, token), ready to render as a diagram.
*_trace_next_hop — "where did the bulk of the funds go": outgoing flow aggregated by counterparty, largest first.
*_trace_dominant_path — auto-follows the largest edge hop by hop, up to 5 hops in a single call.
btc_tx_flow — full inputs and outputs of a Bitcoin transaction with change outputs flagged, so you follow the real payment.
address_labels / addresses_by_label — the label directory both directions: identify one address, or pull every address tagged mixer, sanctioned or scam. A single major exchange resolves to hundreds of thousands of labeled deposit addresses.

Investigate with FIFO, LIFO, or any methodology you prefer — the agent applies it across the raw transfer data. Transfer tracing covers:

EthereumPolygonArbitrumBaseOptimismTronBitcoinSolana

If you work in compliance and AML, investigate fraud, or just follow the money on-chain, this gets you from an address to the whole story in about a minute.

Trading & Market Data

Market intelligence, on demand

Bitquery indexes every DEX trade on Ethereum, Arbitrum, Base, Polygon, Optimism, BNB Chain, Tron and Solana down to the second — so the answers cover freshly-launched memecoins and long-tail tokens that never appear on CoinGecko or CoinMarketCap. You ask in plain English; the agent picks the right tool:

8 chainsof DEX trades indexed

1 secondfinest candle interval

30 daysof per-second trades via SQL

100+MCP tools in the catalog

"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 in the last 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

"Which wallets are quietly accumulating this token right now — buys far above sells?"uses accumulating_traders_by_token; profitable_traders_by_token ranks by realized + unrealized PnL instead

"Decode this transaction — every swap leg, 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)

Guided trading prompts

token_chart

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

token_dashboard

360° view of one token: price chart + supply / FDV + top & smart-money traders + scam check.

trader_dashboard

One wallet: profile stats + holdings / PnL positions + activity time-series.

compare_assets

Overlay 2–5 assets on one normalized %-change price chart + stats table.

token_scout

Trending / gainers + momentum + accumulation → ranked pivot of promising tokens.

market_movers

Chain leaderboard: top gainers + most-traded → sortable table + bar chart.

arbitrage_scan

One asset across chains → price-spread % table + bar chart.

100+ tools under the hood

Every prompt and example above is built from individual MCP tools your agent can also call directly. The catalog groups into six families:

Discovery

find_tokens resolves a name or ticker (with wildcards) to contract + 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 / token_ohlcv, pair_ohlcv (USD- or quote-denominated), cross-chain currency_* aggregates for majors, plus supply, market cap and pool_recent_trades. Candles from 1 second to 1 hour.

Traders

trending_tokens (volume / gainers / losers / most-volatile), top_traders by token, pair or network, profitable_ and accumulating_traders, and per-wallet trader_profile / positions / activity.

Tracing

A per-chain family for 8 chains: *_address_flow_summary, *_flow_edges, *_trace_next_hop, *_trace_dominant_path, *_transfers_in/out, *_tx_transfers, *_token_holders, plus contract *_find_events / *_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 & 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).

The full tool list, parameter shapes, and supported networks are returned by the server itself when your client calls tools/list — so the AI always sees the up-to-date catalog without you reading docs. Per-chain coverage varies slightly (e.g. Solana exposes solana_find_instructions, Bitcoin adds btc_related_addresses clustering); chain_capabilities is the authoritative matrix.

Why use the MCP endpoint instead of GraphQL?

The underlying dataset is the same Bitquery you already know. The difference is the interface:

No query writing. The agent picks the tool and supplies arguments. You stay in conversation.
Structured tool definitions. Each tool advertises its inputs and outputs to the model, so the model can chain calls — for example, find a token, then pull its top traders, then look up each trader's recent positions, in one turn.
Same auth, same plan. Quotas, rate limits, and billing come from your existing Bitquery account. There's no separate signup.
Streaming transport. The server speaks Streamable HTTP, so long responses (large trade histories, multi-step tool chains) stream back without timing out.

How to install

Install Bitquery as a custom connector in your AI client. No config file editing required. Watch the walkthrough:

Claude (Desktop & Web)

Open Settings → Connectors
Click Add custom connector
Enter:
- Name: Bitquery
- URL: https://mcp.bitquery.io
Click Add — a browser window opens to log in with your Bitquery account

Claude Code (CLI)

Add Bitquery to Claude Code with one command (the Add to Claude Code button above copies it to your clipboard):

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

Append --scope user to enable it across every project. On first use, run /mcp inside Claude Code (or claude mcp login bitquery) and complete the browser OAuth login with your Bitquery account.

ChatGPT

Open Settings → Connectors (requires Plus/Pro/Business plan)
Click Add connector → Custom connector
Enter URL: https://mcp.bitquery.io
Authenticate with your Bitquery account

Cursor

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

OpenAI Codex (CLI)

Add Bitquery to your Codex MCP servers with the CLI:

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

Or edit ~/.codex/config.toml directly:

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

On first call, Codex opens a browser for OAuth login with your Bitquery account.

VS Code / other MCP clients

Add a new MCP server with URL: https://mcp.bitquery.io

FAQ

What is the Bitquery MCP Server?

It is a hosted Model Context Protocol (MCP) server at https://mcp.bitquery.io that exposes Bitquery's on-chain blockchain data as MCP tools. AI agents like Claude, ChatGPT, Cursor, and VS Code can call these tools directly during a conversation, without you having to write GraphQL queries by hand.

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 / 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, Polygon, Arbitrum, Base, Optimism, Tron, Bitcoin and Solana, backed by millions of address labels.

Which AI clients can connect to it?

Any client that supports remote MCP servers over Streamable HTTP. This includes Claude Desktop and Claude.ai (web, with Plus/Pro/Max), ChatGPT (Plus/Pro/Business with custom connectors enabled), Cursor, VS Code with MCP enabled, and custom agents built with the Anthropic or OpenAI SDKs.

Which blockchains and datasets are available?

DEX trading tools cover Ethereum, Arbitrum, Base, Polygon, Optimism, BNB Chain, Tron and Solana, with candles down to 1-second granularity; transfer-tracing tools cover Ethereum, Polygon, Arbitrum, Base, Optimism, Tron, Bitcoin and Solana. Tools span token prices, OHLCV candles, trending tokens, top / profitable / accumulating traders, trader profiles and positions, DEX pairs, transfers, the address-label directory, and raw SQL access where the user has a paid plan.

How does authentication work?

Authentication uses OAuth 2.1: the AI client opens a browser to your Bitquery account, you approve access, and tokens refresh automatically for ~30 days. Access is tied to your Bitquery account, so revoking it there immediately disables the connection.

How is this 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_by_token) and the server runs the underlying query on Bitquery for you. The data source is the same; the interface is conversational. Quotas, plans, and billing are shared with your regular Bitquery account.

Is it free?

There is a free tier inherited from your Bitquery account. Heavier workloads — high request rates, large historical scans, or execute_sql — require a paid Bitquery plan. The MCP server itself does not add a separate charge.

Bitquery MCP Server MCP