mnemox-ai

mnemox-ai / tradememory-protocol

Public

MCP server for AI trading memory — 3-layer memory system with LLM-powered reflection engine.

mnemox.ai ai ai-agent algorithmic-trading alpaca binance
62
11
100% credibility
Found Mar 01, 2026 at 53 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

TradeMemory Protocol provides persistent memory for AI trading agents by recording trade decisions, analyzing patterns through reflection, and applying learned adjustments across sessions.

How It Works

1
🔍 Discover TradeMemory

You hear about a helpful tool that lets your trading bot remember past trades and learn from them, just like a journal for smart decisions.

2
🧪 Try the Quick Demo

You play with a simple demo that simulates 30 trades, showing how the bot spots winning patterns and fixes mistakes right before your eyes.

3
✨ See the Magic Happen

Watch in amazement as it analyzes sessions, strategies, and confidence levels, turning random trades into smart adjustments that boost profits.

4
🔗 Connect Your Trading Bot

You easily link it to your AI trading helper so it can save and recall memories across every new trading day.

5
📝 Log Your Real Trades

As your bot makes trades, it automatically records details like reasons, outcomes, and market moments for future learning.

6
💡 Review Daily Insights

Each evening, you check fun summaries of patterns found, like which times or strategies work best, helping you tweak for tomorrow.

🚀 Your Bot Evolves

Over weeks, your trading bot remembers lessons, avoids old mistakes, and grows smarter, leading to more wins and steady growth.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 53 to 62 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is tradememory-protocol?

TradeMemory Protocol is an MCP server that adds persistent memory to stateless AI trading agents. It records every trade decision with reasoning, confidence, market context, and outcomes, then runs a reflection engine to analyze patterns across a 3-layer memory system—active trades, discovered insights, and SQLite-stored strategy adjustments. Built in Python, it integrates via uvx for MCP clients like Claude Desktop or Cursor, with REST API, Docker support, and MT5 sync for real trades.

Why is it gaining traction?

In the MCP ecosystem, this stands out as a specialized Python MCP server for AI trading, compatible with GitHub Copilot VSCode extensions, n8n automations, and project managers via its Streamlit dashboard. Developers grab it for the no-setup demo simulating 30 trades, daily/weekly reflection scripts, and adaptive risk checks that load learned constraints on startup. The 3-layer engine delivers cross-session learning without multi-agent complexity.

Who should use this?

Quant devs building Claude-powered trading bots need this MCP GitHub server to persist agent memory and avoid repeating mistakes. MT5 users automating syncs via Python scripts, or teams evaluating strategies through the dashboard and API endpoints like /reflect/run_daily. Ideal for AI traders in Cursor or GitHub Codespaces wanting MCP server examples and tutorials without crypto/stock distractions yet.

Verdict

Grab it if you're prototyping AI trading agents—solid docs, 181 passing tests, and easy MCP server Docker setup make alpha status (12 stars, 1.0% credibility) forgivable. Skip for production until Phase 2 multi-agent support lands in 2026; otherwise, uvx it for quick memory wins.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.