Practical overview
When it fits
Use it when an AI trading workflow needs to retain trade outcomes, retrieve relevant history before a new decision, or record and evaluate order requests through an MCP-based setup.
Who it is for
- Developers integrating trading memory into AI agents that use MCP.
- Traders who want an agent to consult its recorded trade history.
- Builders seeking a decision-audit trail for agent trading activity.
Documented uses
- Sync supported trading fills into persistent memory.
- Bring prior losses and similar trade outcomes into an agent's context before a proposed trade.
- Record agent decisions in a tamper-evident audit trail.
- Evaluate broker order requests against configured rules through the documented preview proxy.
Requirements & limits
The project is documented as an open-source repository under the MIT license. The README provides local installation and MCP server setup instructions.
- The README names Hyperliquid public fills, Alpaca sync using user-held keys, and MT5 and Binance spot sync scripts.
- Hyperliquid sync covers recent fills rather than a complete address history.
- The described proxy supports Alpaca and requires the proxy extra on Python 3.12 or later.
- The project provides MCP tools; a specific client integration is not documented in the cited setup.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Connect Tradememory Protocol to Claude Desktop
Add the Tradememory MCP server entry to Claude Desktop's configuration.
Before you start
- Python package installation with pip
- Claude Desktop and access to its claude_desktop_config.json file
- uvx
Steps
Install the package with pip install tradememory-protocol.
Add an mcpServers entry named tradememory to claude_desktop_config.json.
Set the entry's command to uvx and its argument to tradememory-protocol.
Expected outcome
Claude Desktop is configured with the Tradememory MCP server entry.
Guide limitations
- The cited setup instructions do not document a specific client-integration test.
Import public Hyperliquid fills
Sync an address's available public fills into local trade memory.
Before you start
- Python package installed with pip
- Hyperliquid address
Steps
Install the package with pip install tradememory-protocol.
Run tradememory sync hyperliquid --address 0xYourAddress, replacing the example with the address to sync.
Expected outcome
Closed trades from the available fills are stored in memory, and the command reports descriptive statistics about the trade history.
Guide limitations
- Hyperliquid's API provides only an address's recent fills, not its full history.
- The statistics describe past trades and are not advice about future trades.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 12 October 2026
- Original website title
- mnemox-ai/tradememory-protocol · Official homepage title · retained in its source language