Skip to main content
Context Protocol Hero
The insight you pay $500/year for? Your agent could buy it for $0.10/response.

Why use Context?

  • 🔌 One Interface, Everything: Stop integrating APIs one by one. Use a single SDK to access any tool in the marketplace.
  • 🧠 Zero-Ops: We’re a gateway to the best MCP tools. Just send the JSON and get the result.
  • ⚡️ Agentic Discovery: Your Agent can search the marketplace at runtime to find tools it didn’t know it needed.
  • 💸 Dual-Surface Economics: Use Query for pay-per-response intelligence or Execute for session-budgeted method calls.

Who Is This SDK For?

RoleWhat You Use
AI Agent Developer@ctxprotocol/sdk (TypeScript) or ctxprotocol (Python) — Query for curated answers or Execute with explicit method pricing + session budgets
Tool Contributor (Data Broker)@modelcontextprotocol/sdk + @ctxprotocol/sdk — Standard MCP server + security middleware
For AI Agent Developers: Use the SDK to search one marketplace with two surfaces:
  • Query surface for curated answers (pay-per-response)
  • Execute surface for deterministic method calls (per-call pricing inside execute sessions)
For Tool Contributors: You need both SDKs:
  • @modelcontextprotocol/sdk — Build your MCP server (tools, schemas, handlers)
  • @ctxprotocol/sdk — Secure your endpoint with createContextMiddleware() to verify requests
Tool contributors wrapping rate-limited APIs: publish _meta.rateLimit hints so Context can pace planner/runtime calls safely. Start with Tool Metadata.

The Ecosystem

Context is not just a chatbot — it’s an economic engine for the AI era. We connect three distinct groups in a decentralized marketplace for Model Context Protocol (MCP) tools. Build a standard MCP server, paste your endpoint URL, set a query price, optionally add method-level execute pricing, and get paid in USDC when agents use your tool.

One Marketplace, Two Surfaces

Listings can expose methods to one or both surfaces using method metadata (_meta.surface):
SurfaceWho is the librarian?Billing modelMethod eligibility
QueryContext app/runtimePay-per-responsesurface is answer or both and queryEligible=true
ExecuteYour app/agentPer execute call (session-budgeted)Explicit method execute pricing (_meta.pricing.executeUsd)
Mixed listings are supported immediately. A listing can have some methods for Query and others for Execute. Methods without explicit execute pricing remain query-only.

For Users: The “God Mode” Agent

Stop switching tabs. Context gives you an AI agent with real-time, read/write access to the world.

Auto Mode

Ask “Find the best yield on Base” and the Agent autonomously searches the marketplace, buys the necessary data tools, analyzes the returns, and presents the answer.

One Wallet

Pay for everything — from gas data to premium stock analysis — using a single USDC balance. No subscriptions, just pay-per-response.

For Tool Builders: The “App Store” for MCP

Monetize your data without building a frontend. Turn the insights people pay $500/year for into $0.10/response revenue you keep.
BenefitDescription
Build Once, Sell EverywhereCreate a standard MCP Server
Zero UI RequiredYou provide the API; our Agent provides the interface
Instant RevenueSet a price (e.g., $0.01/response). Get paid in USDC on Base every time an Agent successfully uses your tool

For App Developers: The Universal Adapter

Build your own agents using the Context Protocol as your infrastructure layer.
1

Stop Integrating APIs

Instead of integrating CoinGecko, Tavily, and Weather APIs separately, just install @ctxprotocol/sdk (TypeScript) or ctxprotocol (Python).
2

Dynamic Discovery

Your agent can search our marketplace at runtime: client.discovery.search("gas prices").
3

Schema Inspection

The API exposes full Input/Output schemas, allowing your LLM to self-construct arguments and execute tools it has never seen before.
4

Zero-Ops

We host the connections. You just send JSON and get results.

The npm/pip for AI Capabilities

“Context Protocol is npm/pip for AI capabilities.”Just as npm install or pip install gives you code other people wrote, Context gives your Agent capabilities other people built.
import { ContextClient } from "@ctxprotocol/sdk";

const client = new ContextClient({ apiKey: "sk_live_..." });

// Query surface: curated intelligence, pay-per-response
const answer = await client.query.run("What are the top whale movements on Base?");

// Execute surface: explicit execute pricing + session budget
const executeTools = await client.discovery.search({
  query: "ethereum gas prices",
  mode: "execute",
  surface: "execute",
  requireExecutePricing: true,
});

const session = await client.tools.startSession({ maxSpendUsd: "2.00" });
const method = executeTools[0]?.mcpTools?.[0];
if (!method) throw new Error("No execute method available");

const result = await client.tools.execute({
  toolId: executeTools[0].id,
  toolName: method.name,
  args: { chainId: 1 },
  sessionId: session.session.sessionId ?? undefined,
});
console.log(result.session); // { methodPrice, spent, remaining, maxSpend, ... }