Skip to content
AI Rook logo

AI Rook MCP server

Which MCP servers are AI Rook'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 2026-10-24

← Back to AI Rook

Whose server it is
  • Registry namecom.ai-rook.agents/trading-mcpverified in the MCP registry

The registry name sits under ai-rook.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-10-24
  • Stateanswerschecked 2026-10-24
  • Tools10
  • Server calls itselfai-rook
  • Server version4.1.0
  • Protocol version2025-03-26

On 2026-10-24 the server answered initialize and tools/list without a sign-in. First answer in 236 ms.

Should an agent connect itSkip it
  • Its tools do none of the product's use cases we checked: search and summarize papers from PubMed, arXiv, and bioRxiv; verify legal document provenance on-chain, page by page; transcribe and synthesize speech in real time over websocket.
  • Checked on 3 of the 7 days it takes to judge reliability.

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: 3, the first on 2026-10-24
  • INITE MCP scoreafter a weeka score needs 7 daily checks: 3 so far
  • What an agent can do with it0 of 3 use cases
  • Use36 npm downloads a week
  • Answeredmeasuring3 of the 7 daily checks it takes to print a share
  • Median first answer411 ms
  • Space in an agent's context≈ 950 tokensthe definitions of all its tools (10), which an agent carries on every turn
  • Tool definitionsB · 83 of 100
  • Typed results0 of 10 toolsdeclared outputSchema: results come typed, read once
  • Not saying whether it only reads or can destroy: 10 of 10

What an agent can do with it

  • cannot · search and summarize papers from PubMed, arXiv, and bioRxiv
  • cannot · verify legal document provenance on-chain, page by page
  • cannot · transcribe and synthesize speech in real time over websocket

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 offers10 tools · read 2026-10-24
  • list_endpoints

    List the full AI Rook endpoint catalog with prices (52 endpoints: trading intelligence, AI inference via local 456B MoE, blockchain data, dev tools, escrow). START HERE before calling any paid endpoint. Free.

  • get_quote

    Get exact price, input schema, and REAL measured latency (from production paid calls) for any AI Rook endpoint BEFORE paying. Agents are risk-averse spenders — quote first, pay second. Free.

  • escrow_dry_run

    Preview a full ERC-8183 escrow job lifecycle on Base WITHOUT touching the chain: exact contract calls with calldata, gas estimates, money split (98% provider / 2% platform), state transitions, reject/refund alternates. Verified against…

  • sign_receipt

    Generate a cryptographically signed proof-of-spend receipt for any USDC payment tx on Base. Server verifies the on-chain Transfer event, then signs with the AI Rook server wallet (EIP-191, ecrecover-verifiable offline). For agent operator…

  • preview_market_pulse

    FREE taste of the paid /api/market-pulse ($0.05): real BTC price + engine status with CVD/OI/whale fields redacted. Use to evaluate data quality before paying. No payment required.

  • preview_order_flow

    FREE taste of the paid /api/order-flow ($0.25): real price + volume gear with CVD/session levels redacted. Evaluate before paying. Free.

  • preview_trade_idea

    FREE taste of the paid /api/trade-idea ($0.25): strategy name + direction only, entry/stop/target redacted. Evaluate before paying. Free.

  • get_escrow_job

    Read the on-chain state of any ERC-8183 escrow job by ID (Open/Funded/Submitted/Completed/Rejected/Expired, client, provider, budget, deliverable hash). Job 39 is our verified live completion. Free.

  • get_attestation

    Read the on-chain attestation record for a completed/rejected ERC-8183 job — provider address, agentId, approved flag, timestamp. Proof that the RookAttestationHook fired. Free.

  • call_paid_endpoint

    Get payment instructions + live quote for any paid AI Rook endpoint. Returns exact USDC price, x402 payment flow (EIP-3009 sign → retry with X-Payment header), and current measured latency. Does NOT execute the endpoint — use this to…

How to connect

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

{
  "mcpServers": {
    "trading-mcp": {
      "type": "http",
      "url": "https://agents.ai-rook.com/mcp"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http ai-rook https://agents.ai-rook.com/mcp

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://agents.ai-rook.com/mcp as a custom connector in the settings.

Also shipped through: npm

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.