More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    oanda mcp

    1

    Production-grade MCP server for OANDA v20 with built-in circuit breakers, rate limiting, and 37 trading actions.

    $19.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    oanda mcp

    oanda mcp

    Example session with this skill installed

    Check my current EUR/USD position and close it if I am in profit by more than 20 pips.

    • Read your context and instructions
    • Compiled the oanda mcp

    I have accessed your OANDA account to check the current status of your EUR/USD positions.

    Position Analysis and Execution

    InstrumentUnitsSideEntry PriceCurrent PriceP/L (Pips)Action Taken
    EUR_USD1000Long1.085021.08722+22.0Closed

    Execution Details:

    • Order ID: #8921-A
    • Status: Filled
    • Closing Price: 1.08722
    • Realized P/L: +$2.20 USD

    Next steps

    • Verify the realized profit in your OANDA transaction history.
    • Monitor for new entry signals on EUR/USD.
    • Update your daily risk report with this completed trade.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Trading via LLMs often lacks the safety and connectivity required for production environments. This skill provides a production-grade Model Context Protocol (MCP) server that bridges OANDA v20 with AI agents, ensuring execution is backed by circuit breakers and rate limits.

    What it does

    • Execute Trades manages market, limit, and stop orders with automated safety checks.
    • Monitor Risk tracks account balances, margins, and position exposures in real-time.
    • Stream Market Data retrieves live price feeds and historical candlesticks for technical analysis.
    • Manage Positions queries open trades and handles batch closures or modifications.
    • Safety Enforcement implements circuit breakers and rate limiting to prevent API exhaustion and account liquidation.
    • Input Validation runs pre-flight checks on all orders to catch errors before they hit the OANDA API.

    How it works

    1. Install the server via Python or Docker and configure your OANDA API credentials in the .env file.
    2. Start the FastAPI server to expose the 37 available trading actions through the MCP protocol.
    3. Connect your agent by adding the server endpoint to your MCP configuration file.
    4. Issue natural language commands to analyze markets or manage orders while the server handles the underlying JSON-RPC communication.

    Frameworks & tools

    Built with Python 3.10, FastAPI, and the oandapyV20 client. It supports Docker deployments and provides OpenAPI/Swagger documentation for direct API interaction.

    Why this beats prompting it yourself

    Prompting an LLM to write trading code often results in fragile scripts that ignore rate limits or fail during high volatility. This skill provides a resilient middleware layer with 391 tests and built-in circuit breakers that protect your capital from programmatic failures.

    Use cases

    • Automated Execution placing orders based on complex technical indicators identified by the agent.
    • Active Risk Monitoring automatically closing positions when specific loss thresholds are breached.
    • Portfolio Auditing generating summaries of transaction history and trade performance.

    Known limitations

    Requires an active OANDA account. Operates on the OANDA v20 API only. Users must manually confirm orders unless they explicitly instruct their agent otherwise.

    Requires Python 3.10+ or Docker. Limited to OANDA v20 API. Database for audit logging is optional. Practice environment used by default for safety.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 11 days ago

    Requires

    Python
    Git
    OANDA v20 API

    Permissions

    Network Access
    Environment Variables
    Read Files
    Listed11 days ago

    What's inside

    Frequently Asked Questions

    This MCP server acts as a safety-first middleware, providing built-in circuit breakers and rate limiting that standard AI-generated scripts lack, ensuring your account isn't liquidated by runaway execution or API bans.

    The skill is compatible with any AI agent or platform that supports the Model Context Protocol (MCP), such as Claude Desktop or bespoke agents built with Python and the MCP SDK.

    You will receive the complete production-grade source code, a Dockerfile for containerized deployment, pre-configured environment templates, and access to 37 specialized trading actions.

    Yes, the server supports both OANDA 'Practice' (demo) and 'Live' accounts; you simply need to toggle the environment variable in your configuration file before starting the server.

    Setup is straightforward: install the dependencies via Python or Docker, input your OANDA API credentials into the .env file, and add the server's endpoint to your MCP configuration.

    The server performs pre-flight input validation and utilizes the oandapyV20 client to ensure all requests meet OANDA’s specific API schema requirements before execution.

    New in Trading & Investing