Agentoracle MCP

Created By
TKCollective2 months ago
MCP server that connects AI assistants (Claude, Cursor, Windsurf) to AgentOracle's real-time research API. Agents can research any topic and get structured JSON with summary, key facts, sources, and confidence scores. Pay-per-query via x402 protocol — $0.02 USDC on Base mainnet. No API keys needed.
Overview

AgentOracle MCP Server

agentoracle-mcp MCP server

An MCP (Model Context Protocol) server that connects AI assistants like Claude, Cursor, and Windsurf to AgentOracle — a pay-per-query real-time research API for AI agents.

What It Does

This MCP server exposes AgentOracle's research capabilities as tools that any MCP-compatible AI assistant can call:

ToolDescriptionCost
researchReal-time research on any topic — returns structured JSON with summary, key facts, sources, and confidence score$0.02 USDC
deep-researchComprehensive deep analysis with expert-level findings, 10-15 detailed facts, and analysis paragraph$0.10 USDC
check-healthCheck if AgentOracle API is online with service statusFree
get-manifestGet x402 payment manifest with pricing and endpoint detailsFree

Payments are handled via the x402 protocol on Base mainnet using USDC.

Quick Start

Install from npm

npm install -g agentoracle-mcp

Or clone and build from source

git clone https://github.com/TKCollective/agentoracle-mcp.git
cd agentoracle-mcp
npm install
npm run build

Usage with Claude Desktop

Add this to your Claude Desktop MCP config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):

{
  "mcpServers": {
    "agentoracle": {
      "command": "npx",
      "args": ["-y", "agentoracle-mcp"]
    }
  }
}

Or if installed globally:

{
  "mcpServers": {
    "agentoracle": {
      "command": "agentoracle-mcp"
    }
  }
}

Usage with Cursor

Add to your Cursor MCP settings:

{
  "mcpServers": {
    "agentoracle": {
      "command": "npx",
      "args": ["-y", "agentoracle-mcp"]
    }
  }
}

Example

Once connected, you can ask your AI assistant things like:

"Use AgentOracle to research the latest DeFi protocol TVL rankings on Base"

The assistant will call the research tool and return structured results:

{
  "query": "Latest DeFi protocol TVL rankings on Base",
  "tier": "standard",
  "result": {
    "summary": "As of March 2026, Base L2 DeFi TVL has reached...",
    "key_facts": [
      "Aerodrome leads Base DeFi with $3.2B TVL",
      "Aave v3 on Base holds $1.8B"
    ],
    "sources": ["https://defillama.com/..."],
    "confidence_score": 0.89
  },
  "confidence": {
    "score": 0.89,
    "level": "high",
    "sources_count": 6,
    "facts_count": 5
  },
  "freshness": "recent",
  "metadata": {
    "model": "sonar",
    "api_version": "1.3.0",
    "response_time_ms": 2100
  }
}

How Payment Works

AgentOracle uses the x402 protocol for payments:

  1. The MCP tool sends your query to AgentOracle
  2. If no payment header is present, the API returns HTTP 402 with payment requirements
  3. Use an x402-compatible client to sign a USDC payment on Base
  4. Resubmit the request with the payment header to get full results

The free check-health and get-manifest tools work without any payment.

Resources

The server also provides an embedded API documentation resource at docs://agentoracle/api that MCP clients can access for integration details.

Tech Stack

License

MIT

Project Info
Created At
2 months ago
Updated At
2 months ago
Author Name
TKCollective
Star
-
Language
-
License
-
Category

Recommend Servers

View All
Tavily Mcp
@tavily-ai

JavaScript
a year ago
AI Work Market — USDC settlement rails for AI labor on Base Mainnet)
@Dario (DME)

AI Work Market is a USDC escrow protocol on Base Mainnet, designed for autonomous AI agents to find work, post jobs, and settle payments without humans in the loop. This MCP server exposes 10 tools: **Escrow lifecycle** - `create_intent_quote` — get calldata + gas estimate for funding a new escrow intent - `submit_proof_quote` — get calldata for the seller to submit a proof URI - `release_funds_quote` — get calldata for the buyer to release payment (or claim/refund) **x402 single-call binding** - `x402_consume` — replaces the 5-step x402 flow with one HMAC-signed POST that returns a delivery URL **Onboarding & discovery** - `agent_onboard` — generate a signed agent card with marketplace attestation - `agent_search` — tf-idf search over the live agent catalog - `agent_reputation` — server-side reputation from on-chain Released/Refunded/Disputed events **Live state** - `system_status` — live on-chain state (nextIntentId, accumulatedFees, contract balance, owner) - `escrow_rules` — contract semantics, lifecycle, call guides, failure modes - `events_subscribe` — SSE stream of new on-chain intent events All endpoints are serverless (Vercel) and return their schema on GET. No browser, no wallet UI required for an agent to integrate. The protocol takes a 1% commission on every settlement; the rest goes to the seller. The full AgentCard is at `/.well-known/agent-card.json` (A2A-compatible). The OpenAPI 3.0.3 spec is at `/.well-known/openapi.json` with `components.securitySchemes` (none, hmacX402). `robots.txt` allows GPTBot, ClaudeBot, anthropic-ai, PerplexityBot, Google-Extended, Applebot-Extended, CCBot, Amazonbot.

16 hours ago
Voyei

6 hours ago