MCP Catalogs
Hometradememory-protocol screenshot

tradememory-protocol

by mnemox-ai·906·Score 54

TradeMemory Protocol is an MCP server providing decision audit trails and persistent memory for AI trading agents with outcome-weighted recall and SHA-256 tamper detection.

financeai-llmdeveloper-tools
119
Forks
1
Open issues
this month
Last commit
2d ago
Indexed

Overview

TradeMemory Protocol serves as a memory layer for AI trading agents, addressing regulatory requirements and performance improvement through comprehensive decision recording and retrieval. The system implements 17 MCP tools across memory, state, planning, risk, and audit categories, enabling agents to recall past trades weighted by outcome quality, record new decisions, and generate tamper-proof audit trails. It supports multiple markets and brokers without executing trades or handling financial assets. The framework includes Outcome-Weighted Memory (OWM) scoring based on cognitive science principles and reinforcement learning research.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:Pre-flight trading workflow: Memory retrieval before making trading decisions
you:Automated trading systems: Sync trade decisions and outcomes from platforms like MT5
you:Compliance and audit: Generate tamper-proof records for regulatory submissions
you:Does TradeMemory execute trades or handle funds?
you:How does it ensure data integrity for compliance?

When to choose this

Choose this when building regulated trading AI systems that require persistent memory, audit trails, and behavioral analysis across trading sessions.

When NOT to choose this

Not suitable if you need a general-purpose memory solution or if you're building simple trading bots without compliance requirements.

Tools this server exposes

12 tools extracted from the README
  • remember_trade

    Record a trading decision with multiple memory layers

  • recall_memories

    Retrieve past trades weighted by outcome quality and context

  • get_agent_state

    Check current trading confidence, drawdown, and streak status

  • get_behavioral_analysis

    Analyze trading patterns and behavioral drift

  • create_trading_plan

    Create a prospective trading plan with conditional triggers

  • check_active_plans

    Review and evaluate active trading plans

  • check_trade_legitimacy

    Evaluate a trade against pre-defined risk factors

  • export_audit_trail

    Export tamper-proof trading decision records

  • verify_audit_hash

    Verify the integrity of a trading decision record

  • get_strategy_performance

    Analyze historical performance of trading strategies

  • get_trade_reflection

    Retrieve detailed reflection notes on past trades

  • validate_strategy

    Validate a trading strategy against historical data

Comparable tools

trading-ai-mcpaudit-trail-mcpmemory-layer-mcp

Installation

pip install tradememory-protocol

Add to Claude Desktop (claude_desktop_config.json):

{
  "mcpServers": {
    "tradememory": {
      "command": "uvx",
      "args": ["tradememory-protocol"]
    }
  }
}

For other clients:

# Claude Code
claude mcp add tradememory -- uvx tradememory-protocol

# From source
git clone https://github.com/mnemox-ai/tradememory-protocol.git
cd tradememory-protocol && pip install -e . && python -m tradememory

# Docker
docker compose up -d

FAQ

Does TradeMemory execute trades or handle funds?
No. TradeMemory only records and recalls trading decisions. It never executes trades, moves funds, or accesses wallets.
How does it ensure data integrity for compliance?
Every trading decision is SHA-256 hashed at creation and linked into a forward-chained audit ledger. Daily Merkle roots anchor the chain, making tampering detectable.

On Hacker News

Recent discussion from the developer community.

Compare tradememory-protocol with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.