Aevo (@aevoxyz): Introducing Aevo MCP and AI Trading Bot

Introducing Aevo MCP and AI Trading Bot MCP servers for trading are proliferating. In crypto, integrations from CoinGecko, 0x-based aggregators and Uniswap give agents access to market data and spot swaps. TradFi platforms have shipped MCP servers with broader capabilities, including options. But across all of DeFi, none of these integrations can construct a multi-leg position, calculate portfolio Greeks, or run pre-trade risk validation; they stop where the complexity starts. That’s exactly where Aevo has always operated. Our mission, since day one, has been to make sophisticated instruments accessible to a wider audience. AI agents are the next step in that arc: traders can now say “build me a delta-neutral position on ETH” or “set up an iron condor with defined risk,” and have agents reason through the strategy, validate the risk, and execute it. The complexity becomes something you can delegate, while you focus on your desired outcomes. Today we’re launching two new products that make this real: The Aevo MCP Server, an open-source integration that gives any AI agent full access to the Aevo exchange through 45+ structured tools, covering everything from market analysis to multi-leg options execution. And the Aevo Trading Bot on Telegram, the same capability in a chat window, powered by Claude, ChatGPT, and other LLM backends, for traders who want AI-assisted trading without any setup. What is MCP? The Model Context Protocol has become the standard way AI agents connect to external tools and data. Thousands of MCP servers exist today for databases, developer tools, enterprise software, and financial services. The Aevo MCP Server is the first DeFi-native implementation built for derivatives complexity (perpetuals and options), not just spot trading. 45+ Tools for the full trading lifecycle The server exposes over 45 tools organized across five categories that cover everything a trading agent needs: Market Analysis: real-time prices, funding rates, volatility snapshots, and market regime classification. Your agent can ask “what’s the current funding rate on ETH-PERP?” and get a structured answer it can reason about. Portfolio Management: Positions, risk metrics, Greeks exposure, and margin monitoring, so your agent always knows where it stands. Order Execution: Limit orders, bracket orders with entry + stop-loss + take-profit, and batch operations; not just single-shot trades but structured execution plans. Options Strategies: Straddles, strangles, spreads, iron condors, and butterflies with live pricing: multi-leg strategies can be constructed and validated in seconds. Risk Management: Pre-trade validation, portfolio risk scoring, and liquidation distance monitoring. The server also ships with built-in prompts, like "trade_plan", "risk_checklist", and "onboarding_plan", that give your agent structured workflows instead of leaving it to figure out sequencing on its own. Works everywhere, deploys anywhere We built the MCP Server to meet developers where they are: Hosted endpoint: point any MCP client to https://mcp.aevo.xyz/mcp and start trading. No local setup, authenticate with your API key and you’re live. Self-hosted via Docker: clone the repo, add your credentials, "docker compose up". Local install: "pip install" and run via stdio for the tightest integration loop. It works with Claude Desktop, Claude Code, Cursor, Windsurf, OpenClaw, and any other MCP-compatible client: the server is fully open-source on GitHub. There’s also a companion Trading Skills repo that provides structured guidance for agents: instrument naming conventions, risk guardrails, workflow recipes, and an options strategy reference, like a trader’s playbook that the AI can internalize. Bringing it to Telegram Not everyone lives in a code editor: millions of crypto traders spend their days in Telegram, monitoring channels, discussing positions, reacting to news. So we built a Telegram bot that brings agentic trading directly into that environment. The Aevo Trading Bot connects to the same Aevo MCP infrastructure and supports multiple LLM backends, including Claude, ChatGPT, and others. You can interact with it in natural language: ask it to analyze a market, check your positions, or walk you through a trade plan. It’s the same 45+ tool capability set, accessible from a chat window. Get started The Aevo MCP Server is live now. Here’s how to try it: Get your API credentials for your Aevo account Connect to the hosted endpoint at https://mcp.aevo.xyz/mcp, or clone the repo and self-host For risk-free testing, use the testnet at https://mcp-testnet.aevo.xyz/mcp Or skip the setup entirely and open the Aevo Trading Bot on Telegram Full documentation, the server source code, and the trading skills repo are all open-source: Aevo MCP Server on GitHub Aevo Trading Skills on GitHub Aevo API Documentation Aevo Trading Bot on Telegram Aevo MCP Official Docs