Skip to content
Thru logo

Thru MCP server

Which MCP servers are Thru's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.

read 2026-10-21

← Back to Thru

Whose server it is
  • Registry nameorg.thru/thru-explorerverified in the MCP registry

The registry name sits under thru.org. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-10-21
  • Stateanswerschecked 2026-10-21
  • Tools8
  • Server calls itselfthru-explorer
  • Server version0.1.0
  • Protocol version2025-06-18

On 2026-10-21 the server answered initialize and tools/list without a sign-in. First answer in 133 ms.

Should an agent connect itSkip it
  • Its tools do none of the product's use cases we checked: author blockchain programs in C, C++ or Rust; install the thru command-line tool via npm.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

Measured dailydaily checks: 15, the first on 2026-10-21
  • INITE MCP score68 of 100answers 100 · does the job 0 · definitions 83 · context 100 · connects 100
  • What an agent can do with it0 of 2 use cases
  • Answered100% of days15 of 15 daily checks
  • Median first answer135 ms
  • Space in an agent's context≈ 671 tokensthe definitions of all its tools (8), which an agent carries on every turn
  • Tool definitionsB · 83 of 100
  • Typed results0 of 8 toolsdeclared outputSchema: results come typed, read once
  • Not saying whether it only reads or can destroy: 8 of 8

What an agent can do with it

  • cannot · author blockchain programs in C, C++ or Rust
  • cannot · install the thru command-line tool via npm
  • not sure · trace events and track compute usage in programs

Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?

Worth a look

None of the five warning signs in these definitions.

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

Tools it offers8 tools · read 2026-10-21
  • get_block

    Get block details by slot number, including producer, timestamp, and transactions

  • get_transaction

    Get transaction details by signature, including status, accounts, instructions, and events

  • get_account

    Get account details by address, including balance, owner, and metadata

  • list_account_transactions

    List transactions for an account with pagination

  • list_recent_blocks

    List the most recent blocks on the chain

  • list_recent_transactions

    List the most recent transactions across recent blocks

  • search

    Search for blocks, transactions, or accounts by query (block number, tx signature, or address)

  • get_program_abi

    Get the on-chain ABI for a program, including instruction and event definitions

How to connect

Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:

{
  "mcpServers": {
    "thru-explorer": {
      "type": "http",
      "url": "https://scan.thru.org/api/mcp"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http thru https://scan.thru.org/api/mcp

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://scan.thru.org/api/mcp as a custom connector in the settings.

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.