For AI & agent builders

On-chain data your LLM agents can query in plain English.

Give your AI agents a real-time window into 40+ blockchains. Connect over MCP or API, let the model turn natural language into tool calls, and get back decoded, multichain data agents can reason on and act on — no per-chain plumbing.

Connect the MCP server →Talk to sales

Trusted by leading Web3 teams

Trusted by leading Web3 and AI builders

The challenge

AI agents are blind to on-chain reality.

An agent that can't read the chain can't trade, monitor or answer with confidence. Most teams spend weeks wiring up nodes and decoders per chain just to feed one agent — and it breaks the moment a new chain matters.

Agents can't read on-chain data

LLMs have no native access to wallets, tokens, trades or contract state — so they hallucinate or stay silent.

Brittle per-chain integrations

Hand-rolled RPC and decoders per chain break on schema changes and never quite cover the next ecosystem.

Data backends are slow to build

Standing up a real-time, multichain data layer just to power one agent burns weeks you don't have.

How it works

From a user prompt to on-chain action.

Plug Bitquery into your agent over MCP or API and let the model do the querying — we handle indexing, decoding and delivery.

Connect

Wire your agent to Bitquery over the MCP server or GraphQL API — no per-chain setup.

Ask

The LLM turns natural-language intent into structured tool calls and queries.

Retrieve

Get decoded, multichain data back — trades, balances, transfers, prices, labels.

Act

Feed grounded data into reasoning, alerts, trades or downstream automations.

What you get

Everything an on-chain AI agent needs.

Native MCP server

mcp.bitquery.io exposes our GraphQL APIs as native MCP tools, so Claude, Cursor and any MCP client can query chains with zero custom integration.

Natural-language queries

Our schema is built so an LLM can construct a query directly from a user prompt — no intermediate translation layer required.

Real-time + historical

Subscribe to Kafka, WebSocket and Solana gRPC streams for live triggers, or pull deep history for context and backtesting.

40+ chains, one interface

EVM, Solana, Bitcoin, Tron and more behind a single API and schema — agents reason across ecosystems, not one chain.

Structured for LLMs

Clean, decoded JSON with entity labels and attribution agents can reason on directly — no brittle parsing of raw RPC.

Flexible delivery

MCP tools, GraphQL queries, or real-time streams — wrap as a LangChain, OpenAI function-calling or AutoGPT tool in minutes.

Built on Bitquery

The Bitquery products that power it.

Compose the exact data layer your agent needs — each is a production API with its own docs and SLAs.

McMCP Server →Bitquery as native MCP tools for Claude, Cursor and any MCP client.DxDEX Trades API →Real-time and historical DEX trades across chains for trading agents.TpToken Price API →Live and historical token prices to ground agent decisions.DsCloud Datashares →Bulk on-chain history delivered to your warehouse for context and training.WpWallet Portfolio →Multichain balances, positions and history for portfolio agents.CcCross-Chain API →Query and correlate activity across 40+ chains from one interface.

40+

Blockchains covered

Data delay, real-time

1PB+

On-chain data indexed

Plain

English to query

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

Webacy

webacy.com

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.

Alex Knaggs0x 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.

Flora SunDirector, 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.

Jan DreskeBackend 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.

Nick ChristieCo-Founder, Syla

Bitquery's products are very intuitive and easy to use. We rely on them for DEX trading and liquidity data — saving us the manpower and tedious technical details of building our own system. Their team gives near-24/7 support and resolves issues fast.

Data TeamOurbit

FAQ

On-chain data for AI agents, answered.

How can AI agents use Bitquery's blockchain data?

Bitquery is the on-chain data layer for AI agents that need to read, reason about, and act on blockchain state. Agents use our GraphQL API for ad-hoc queries, our streams for real-time triggers, and our entity-attribution data to make informed decisions about wallets, tokens, and contracts.

Does Bitquery have an MCP server for AI agents?

Yes. The Bitquery MCP (Model Context Protocol) server at mcp.bitquery.io exposes our GraphQL APIs as native MCP tools, so Claude, Cursor, and any MCP-compatible LLM client can directly query blockchain data with no custom integration.

Can I integrate Bitquery with LangChain, OpenAI tools, or AutoGPT?

Yes. Bitquery's GraphQL API is straightforward to wrap as a LangChain tool, OpenAI function-calling tool, or AutoGPT plugin. We also have a native MCP server for clients that prefer the standard MCP protocol over custom tool wrappers.

Does Bitquery support real-time data for autonomous trading agents?

Yes. Trading agents subscribe to our Kafka, WebSocket, or Solana gRPC streams (sub-300ms) to react to DEX trades, new token launches, wallet movements, and price events as they happen. Common use cases: copy trading, MEV detection, sniping bots, and market-making.

How do AI agents query blockchain data through Bitquery?

Agents send GraphQL queries with natural-language-derived parameters (token address, wallet, time range, chain) and receive structured JSON responses. Our schema is designed so an LLM can construct queries directly from a user prompt without intermediate translation layers.

Give your agents the on-chain truth.

Connect the Bitquery MCP server to your agent in minutes, or talk to our team about an MCP and API data layer tailored to your use case.

Connect the MCP server →Talk to sales

40+ chains · real-time · mcp.bitquery.io · plain-English queries