> ## Documentation Index > Fetch the complete documentation index at: https://docs.ifalabs.com/llms.txt > Use this file to discover all available pages before exploring further. # What is the IFÁ Labs MCP Server? > Query live stablecoin price feeds directly from AI agents, developer tools, and LLM-powered applications using the IFÁ Labs MCP Server. The IFÁ Labs MCP Server exposes the full IFÁ Labs oracle infrastructure through the [Model Context Protocol (MCP)](https://modelcontextprotocol.io) — an open standard that lets AI agents, coding assistants, and LLM-powered applications call external tools and data sources directly. In practical terms: instead of writing a contract, running a script, or opening Basescan to check a stablecoin price, you can query IFÁ Labs price feeds conversationally — from Claude, Cursor, Windsurf, TRAE, or any MCP-compatible client — and get verified, live on-chain data back in seconds. *** ## What MCP Is Model Context Protocol is an open standard developed by Anthropic that defines how AI models communicate with external tools and data sources. It works like this: 1. An MCP server exposes a set of named tools — functions with defined inputs and outputs. 2. An MCP client — a coding assistant, an AI agent, an LLM application — discovers those tools and can call them during a conversation or workflow. 3. The server executes the tool call, fetches real data, and returns a structured result. 4. The model uses that result to inform its response. The result is AI tooling that works with live, verifiable data — not hallucinated values or stale training knowledge. *** ## What the IFÁ Labs MCP Server Exposes The IFÁ Labs MCP Server provides direct access to the full oracle infrastructure: | Tool | Description | | ----------------------- | ------------------------------------------------------------------------------ | | `get_asset_price` | Fetch the current on-chain price for a single supported asset | | `get_multiple_prices` | Fetch prices for multiple assets in a single call | | `get_derived_pair` | Get a cross-asset price — e.g. CNGN priced in USDT terms | | `get_supported_assets` | List all currently supported assets and their IDs | | `get_asset_id` | Resolve a symbol string to its `bytes32` asset ID | | `check_price_freshness` | Check the age of a price feed and whether it meets a given staleness threshold | | `get_network_info` | Return contract addresses and RPC details for all supported networks | Every tool call reads directly from the deployed on-chain oracle contracts. No cached middleware, no synthetic data, no approximations. *** ## Who It's For **Smart contract developers** using AI coding assistants can query live prices mid-session — confirming what CNGN/USD is trading at right now, verifying an asset ID before hardcoding it, or checking feed freshness before deploying to mainnet — without leaving their editor. **DeFi protocol teams** running automated workflows can integrate IFÁ Labs price feeds into agent-based pipelines — monitoring feeds, triggering alerts, or feeding price data into downstream logic — without building a custom integration from scratch. **AI application developers** building LLM-powered financial tools can ground their applications in verified on-chain stablecoin pricing data, accessed through a standard protocol that works with any MCP-compatible model or framework. **Researchers and analysts** can query historical and live stablecoin price data conversationally, without writing scripts or navigating block explorers. *** ## Supported MCP Clients The IFÁ Labs MCP Server works with any MCP-compatible client: | Client | Type | | ---------------------------------------- | ------------------------------------------------------ | | [Claude](https://claude.ai) | AI assistant — Anthropic | | [Cursor](https://cursor.com) | AI coding assistant | | [Windsurf](https://codeium.com/windsurf) | AI coding assistant | | [TRAE](https://trae.ai) | AI coding assistant | | [Zed](https://zed.dev) | AI-powered editor | | [Continue](https://continue.dev) | Open-source AI coding assistant | | Any MCP-compatible client | Custom agents, LangChain, CrewAI, and other frameworks | *** ## A Practical Example Here's what using the IFÁ Labs MCP Server looks like inside an MCP-compatible coding assistant: **Developer prompt:** > "What is the current CNGN/USD price on Base Mainnet? Is the feed fresh?" **MCP tool calls triggered:** 1. `get_asset_price` → `{ asset: "CNGN/USD", network: "base-mainnet" }` 2. `check_price_freshness` → `{ asset: "CNGN/USD", max_age_seconds: 3600 }` **Result returned to the model:** ```json theme={null} { "asset": "CNGN/USD", "price": "0.000613", "raw_price": "613000000000000", "decimal": -18, "last_updated": "2025-07-31T14:22:11Z", "age_seconds": 412, "is_fresh": true, "network": "base-mainnet" } ``` **Model response to developer:** > The CNGN/USD price on Base Mainnet is currently \$0.000613. The feed was last updated 412 seconds ago and is within the 1-hour freshness threshold. No scripts. No block explorer. No manual RPC calls. Live on-chain data, in the flow of a development conversation. *** ## Why This Matters for Stablecoin Infrastructure Most oracle data accessible via AI tools today comes from training data — which has a cutoff date and cannot reflect current market conditions — or from general-purpose web searches that may pull from unreliable sources. The IFÁ Labs MCP Server changes that for stablecoin pricing in emerging markets. It gives any MCP-compatible tool access to the same verified, manipulation-resistant, on-chain price data that production DeFi protocols rely on — with the same trust model, accessed through a standardised interface. For developers building on African, Latin American, or Southeast Asian stablecoins, this means AI tooling that actually knows what CNGN, ZARP, and BRZ are worth right now. *** ## Next Steps Connect the IFÁ Labs MCP Server to your preferred client in minutes. Full reference for every tool the MCP server exposes.