Hi HN, I built this because I trade XAUUSD with AI agents and kept hitting the same wall: the AI forgets everything between sessions.
TradeMemory Protocol is a 3-layer memory system (raw trades → pattern discovery → strategy evolution) exposed as MCP tools. Any MCP-compatible AI agent can plug in.
Stack: Python, FastAPI, SQLite, Claude API for the reflection engine. 36 tests passing, MIT licensed.
My own quant system is the first production user. Would love feedback from anyone building AI agents or doing algo trading.
Happy to answer any questions about the architecture or MCP integration.
Hi HN, I built this because I trade XAUUSD with AI agents and kept hitting the same wall: the AI forgets everything between sessions.
TradeMemory Protocol is a 3-layer memory system (raw trades → pattern discovery → strategy evolution) exposed as MCP tools. Any MCP-compatible AI agent can plug in.
Stack: Python, FastAPI, SQLite, Claude API for the reflection engine. 36 tests passing, MIT licensed.
My own quant system is the first production user. Would love feedback from anyone building AI agents or doing algo trading.
Happy to answer any questions about the architecture or MCP integration.