Resources · Claude connector

Claude Blockscout connectorWhat Claude can do in your Blockscout account.

The Claude Blockscout connector exposes 15 tools. 12 read on-chain data such as balances, tokens, NFTs, transactions and contracts, 0 write anything, and 3 are not described by any source. Below: what Claude can look up, and the key question that decides how well it works.

Verified Trustpilot reviews · AI, automation & growth agency

Overview

What changes when Claude can query Blockscout

Answering a question about a wallet or a contract usually means opening a block explorer, pasting an address, switching tabs between transactions, tokens and logs, and adding things up by hand. With Blockscout connected, you ask Claude in plain words and it pulls the data from the explorer across many chains, then reasons over it in the conversation.

Size up an address in one question. get_address_info returns balance, ENS name, contract status and public tags, and get_tokens_by_address lists the ERC-20 holdings with market data.

Follow the money over a period. get_transactions_by_address and get_token_transfers_by_address work within a time range, which is how Claude can answer a question about one given month of activity.

Look inside a contract. get_contract_abi fetches its interface and read_contract runs a read-only function to return a live value, such as whether a token approval is set for a given spender, without you opening the explorer at all.

What it will not do: no documented tool signs or broadcasts a transaction, and nothing watches a wallet on its own; Claude reads only while it answers. Three tools on the directory sheet have no description anywhere. For the other families covered on this site, start from the Integrations hub.

Vocabulary

Five words first

The Claude terms this page uses, explained once each.

Connector
A link you set up once between Claude and an outside service, so Claude can call on it while it answers your questions.
Tool
One named action a connector hands to Claude. Claude picks the ones it needs, and the directory sheet lists every one by name.
Authorization
The service's own screen where a person grants Claude the access it will use afterwards. It is given once and can be withdrawn.
Approval
The confirmation Claude waits for before doing something that changes an account, shown in the conversation when it matters.
MCP
The shared standard connectors are built on: it is what lets an assistant such as Claude talk to an outside service like Blockscout.
Connect

Connect the Claude Blockscout connector in three steps

  1. 01

    Find Blockscout in the list

    Open Claude's settings, then Customize and Connectors, and look for Blockscout. On a Team or Enterprise workspace, an Owner or Primary Owner must enable the connector for the organization before members can each use it.

  2. 02

    Start the connection

    Click Connect on the Blockscout row and complete whatever the window asks. If the link breaks later, choose Disconnect and connect again, which is the repair Claude documents for a damaged connection.

  3. 03

    Check the authorization

    When a service shows an authorization screen, read it before accepting. It is written by the service, not by Claude, and it sets the scope of the access Claude will rely on.

Tools

The 15 tools, by what is known about them

Blockscout gives Claude 15 tools: 12 that read your account, 0 that change something in it, and 3 no official source describes.

Twelve tools are described in Blockscout's own documentation. Three appear on the directory sheet only, with no description. Names stay exactly as Claude displays them.

  • 12 read
  • 3 not documented

What Claude reads (12)

12 tools

Twelve tools that look up chains, addresses, tokens, blocks, transactions and contracts without touching anything.

__unlock_blockchain_analysis__

Opens a Blockscout session for Claude and hands back reference data and usage rules; Blockscout asks that it run once per session, before any other tool.

When it helps
it is the silent first call behind any blockchain question you ask.

Sourcegithub.com · October 1, 2026 ↗

get_chains_list

Returns the chains Blockscout supports, filterable by name, chain ID, native currency or ecosystem, so Claude knows where it can look.

When it helps
you wonder whether a smaller rollup is covered before spending time asking about it, and want the exact chain ID.

Sourcegithub.com · October 1, 2026 ↗

get_address_by_ens_name

Turns an ENS name into the Ethereum address behind it, the usual first step when a question starts from a readable name rather than a long hexadecimal string.

When it helps
someone asks what a wallet named in a forum post actually holds.

Sourcegithub.com · October 1, 2026 ↗

get_contract_abi

Retrieves a smart contract's ABI, the description of its functions and events, which Claude needs before it can read the contract properly.

When it helps
an auditor wants to know which methods of a contract could emit a given event.

Sourcegithub.com · October 1, 2026 ↗

get_address_info

Gives a full picture of one address: balance, ENS association, whether it is a contract, token details and public tags. The ENS link lets Claude put a readable name next to the hex string.

When it helps
a compliance analyst checks whether an incoming wallet carries a known public label.

Sourcegithub.com · October 1, 2026 ↗

get_tokens_by_address

Lists the ERC-20 tokens an address holds, enriched with metadata and market data, so Claude can put a value next to each holding.

When it helps
a treasury lead wants the token mix of the DAO's main wallet, with current market values, before a governance vote.

Sourcegithub.com · October 1, 2026 ↗

get_transactions_by_address

Pulls the transactions of an address within a time window you set, optionally narrowed to certain methods, so only the calls you care about come back.

When it helps
you want every interaction a wallet had with one protocol during the last quarter.

Sourcegithub.com · October 1, 2026 ↗

get_token_transfers_by_address

Returns the ERC-20 transfers of an address within a time range, so flows in and out can be traced.

When it helps
an accountant reconciles stablecoin payments received by a company wallet in May against the invoices issued that month.

Sourcegithub.com · October 1, 2026 ↗

nft_tokens_by_address

Shows the NFTs an address owns, grouped by collection, so Claude can describe a holding at a glance and point out the larger sets.

When it helps
a collector wants a clean inventory of a wallet, collection by collection, before a sale.

Sourcegithub.com · October 1, 2026 ↗

get_block_info

Reports on one block: timestamp, gas used, burnt fees and transaction count, with the list of transaction hashes if asked.

When it helps
you want to know when a recent block on Gnosis Chain was produced, how much gas it used and how many transactions it held.

Sourcegithub.com · October 1, 2026 ↗

get_transaction_info

Explains one transaction in depth, with decoded input parameters and the token transfers it triggered. Decoded parameters show what was called, with which values, rather than raw hex.

When it helps
a user asks what really happened in a swap that cost more than expected.

Sourcegithub.com · October 1, 2026 ↗

read_contract

Calls a read-only function of a smart contract and returns its result, without changing anything on chain. Blockscout describes it for read-only functions and nothing else.

When it helps
you check whether a token contract includes a blacklisting function and how it answers for one address.

Sourcegithub.com · October 1, 2026 ↗

Not documented (3)

3 tools

Listed on the directory sheet, missing from Blockscout's documentation. What they return is unknown, so they are not explained here.

get_latest_block

No source documents this tool: it is on the directory sheet, but Blockscout's own list of tool descriptions does not include it at all. Its effect is therefore not described here, and the general approval rule further down is what covers it.

transaction_summary

Blockscout's documentation has no entry for this name, and the directory publishes it without a description of any kind. This page therefore states nothing about what it returns; like every tool, it sits under the general approval rule explained below.

get_transaction_logs

Only the name is public: the directory lists it, the publisher's README leaves it out, and no official text says what it does. Rather than guess from the name, the page leaves it undescribed and refers to the general approval rule.

Approvals

What Claude asks before acting

By default, Claude stops and asks for confirmation before each action it takes on an account for someone, in the conversation itself.

The twelve documented tools read public blockchain data, and no source describes a prompt for any Blockscout tool, the three undocumented ones included. On Team and Enterprise workspaces, owners decide whether members may let some actions through without a prompt each time, and they can cap what a connector may do for the whole organization; nobody overrides that from their own account. Claude also works only with the access the connection was given.

Plans

Which plans include Blockscout

No official source publishes plan availability connector by connector: 0 of the 819 directory sheets shows it, and the gap is real.

The published rule is general. Remote connectors are open to all users on Claude, Cowork, Claude Desktop and mobile; desktop extensions install on Claude Desktop. On Team and Enterprise, an Owner or Primary Owner opens a connector for the organization before members connect. The Blockscout sheet in the official directory shows its current state. Only Blockscout documents this connector; no Claude help page covers it.

Limits

Where the Blockscout connector stops

A connector is not an automation. Claude calls these tools while it answers you, so no alert fires when a watched wallet moves funds.

By default, the sheet exposes only reading among documented tools, an observed floor rather than a guarantee: an administrator can open actions no public sheet shows. The partner badge is not a security audit, and Anthropic writes on every sheet that it neither chooses a publisher's tools nor guarantees their behavior. Connect what comes from a publisher you know. For a connector that works on a private business account, compare with the Claude adobe-workfront connector.

Need help

Need help connecting Blockscout to Claude?

A person reads every message.

FAQ

Questions about the Claude Blockscout connector

01What can Claude do with the Blockscout connector?
Claude can read public blockchain data across many chains and explain it to you. Twelve documented tools cover supported chains, ENS lookups, address overviews, ERC-20 holdings, NFTs, transactions and token transfers over a time range, block and transaction details, contract ABIs and read-only contract calls. Three more tools are listed on the directory sheet without any description, so what they return is not known. Claude picks the tools itself as it answers.
02Can Claude send transactions or move funds with Blockscout?
No documented tool does that. Blockscout describes its tools as data access, and even the contract tool is explicitly a read-only function call. By default, the sheet therefore exposes only reading among documented tools, which is an observed floor and not a promise, since three tools remain undescribed. Nothing in the sources mentions signing, a wallet connection or broadcasting a transaction from the chat. If a write tool ever appears on the sheet, this answer would change.
03Does Claude ask before using the Blockscout tools?
By default, Claude asks for confirmation before each action it takes on an account for someone. No source describes a specific prompt for any Blockscout tool, so that general rule is what covers them, including the three undocumented ones. Claude also works only with the access the connection was given. In an organization workspace, owners decide whether members may let some actions pass without a prompt each time, and they can restrict what the connector may do for everyone.
04Which Claude plans is the Blockscout connector on?
No official source publishes plan availability for one connector, and none of the 819 directory sheets lists it. The general rule says remote connectors are open to all users on Claude, Cowork, Claude Desktop and mobile, and that on Team and Enterprise an Owner or Primary Owner enables a connector before members use it. The Blockscout sheet in the directory shows the current state for your account: it is the only place that does.
05Do I need a Blockscout API key to use the connector?
It depends on how you add it. Blockscout says most data tools route through its PRO API and need a key. Installed from the directory, the connector runs on a shared key, which may restrict access and capabilities. Added as a custom connector, it can carry your own key in a request header, a feature still in beta for some organizations. Without any key at all, Blockscout says the data tools fail before reaching its servers.
06Why does the directory list tools Blockscout does not document?
The two lists come from different places and do not match. The directory sheet shows 15 tools; Blockscout's README describes 16, including four absent from the sheet, while three names on the sheet appear nowhere in the README. This page follows the sheet, which is what Claude actually exposes, and leaves those three tools undescribed rather than guessing what they do from their names. If Blockscout documents them later, the page can describe them then.
07Claude or an automation tool for Blockscout data?
They serve different needs, so the choice follows the task rather than a ranking. Claude suits investigation in conversation: ask about a wallet, then a transaction, then a contract, and keep the thread. An automation tool runs a fixed sequence without anyone watching, for example on every new block, which this connector does not do. It has no trigger, and Claude reads data only while it answers. For one-off research, the conversation is usually faster.