Skip to content

feat: TradeMemory Bridge — persistent memory for AI trading agents#108

Open
zychenpeng wants to merge 1 commit intobinance:mainfrom
zychenpeng:feature/tradememory-bridge
Open

feat: TradeMemory Bridge — persistent memory for AI trading agents#108
zychenpeng wants to merge 1 commit intobinance:mainfrom
zychenpeng:feature/tradememory-bridge

Conversation

@zychenpeng
Copy link
Copy Markdown

What This Skill Does

TradeMemory Bridge connects Binance spot trading events to TradeMemory Protocol, giving AI trading agents persistent, outcome-weighted memory.

Capabilities

  • Trade Journaling — Store every Binance spot trade with full market context
  • Outcome-Weighted Recall — Before entering a new position, recall past trades in similar market conditions, ranked by outcome quality
  • Behavioral Bias Detection — Detect overtrading, revenge trading, and disposition effect from trade history
  • Strategy Performance — Track win rate, profit factor, and aggregate stats per strategy
  • Conditional Trading Plans — Set rules that trigger on specific market conditions
  • Agent State Monitoring — Track confidence, drawdown, and win/loss streaks

How It Works

The skill documents 10 MCP tools provided by TradeMemory Protocol. An AI agent reads the SKILL.md, then calls the MCP tools to store and recall trades.

Binance Spot Skill (execute trade) → TradeMemory Bridge (remember_trade) → Persistent Memory
                                   ← TradeMemory Bridge (recall_memories) ← Before next trade

Technical Details

  • Protocol: MCP (Model Context Protocol) via FastMCP
  • Storage: Local SQLite by default (no external data transmission)
  • Tests: 399 passing
  • License: MIT
  • PyPI: pip install tradememory-protocol

Links

…ents

Adds a bridge skill that connects Binance spot trading events to
TradeMemory Protocol (MCP server). Enables AI trading agents to:

- Journal trades with full market context
- Recall similar past trades before entering new positions
- Detect behavioral biases (overtrading, revenge trading)
- Track strategy performance across sessions

Uses 10 MCP tools via TradeMemory Protocol (MIT, 399 tests).
GitHub: https://github.com/mnemox-ai/tradememory-protocol
@theo-s68 theo-s68 self-assigned this Apr 15, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants