MCP MCP server · io.github.mnemox-ai
io.github.mnemox-ai/tradememory-protocol
Tamper-evident decision audit trail and outcome-weighted memory for AI trading agents.
- host stars
- 1,418
- 30-day movement
- +10/day
- Related entries
- 8
- Connections
- 20
pypilocalstdio
TradeMemory is an MCP server giving AI trading agents a tamper-evident audit trail and outcome-weighted memory of past decisions. It records Trading Decision Records (including decisions not to trade) with SHA-256 hashes, and provides 20 tools covering memory recall, plan checks, strategy analysis, and an evolution workflow (market data, pattern discovery, backtests, strategy evolution).
You want your trading agent to recall past outcomes before acting and to produce verifiable decision logs for compliance or review.
Use it to
- Recall how similar market situations played out before a trade
- Log and hash every trading decision for audit
- Verify audit trail integrity with hash-chain checks
- Run pattern discovery and backtests to evolve strategies
- Analyze strategy performance and compute decision quality
For Developers of AI trading agents and compliance-focused trading teams
- Runs
- installed — on your machine
- Transport
- stdio
- Tools
- 20
- Packaged as
- pypi:tradememory-protocol
- Install
- uvx tradememory-protocol
- Version
- 0.5.5
- Last release
- 2026-09-08
topicsmcp-servertradingaudit-trailmemorybacktestingself-hosted