# mnemox-ai/tradememory-protocol

Decision audit trail + persistent memory for AI trading agents. Outcome-weighted recall, tamper-evident SHA-256 chain with RFC 3161 anchoring, 20 MCP tools.

Repository: https://github.com/mnemox-ai/tradememory-protocol
Canonical: https://ross.abutalabs.com/products/tradememory-protocol
Homepage: https://mnemox.ai/tradememory/
Language: Python
License: MIT
License Family: permissive
Topics: claude, forex, mcp, memory, mt5, trading, ai-agents, crypto, evolution-engine, mcp-server, outcome-weighted-memory, agentic-trading, audit-trail, compliance
Last push: 2026-08-11T17:55:53+00:00

## Health v2 (maintenance only)
Score: 80/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 97, release rhythm 95, longevity 13
- inputs: {"age_days": 191, "days_push": 22, "days_rel": 36, "gap_med": 2, "n_releases_24m": 10}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1412, forks 166 (observed 2026-08-28T04:04:39.160571+00:00)

## What it is
TradeMemory Protocol is an MCP server providing persistent memory and a tamper-evident decision audit trail for AI trading agents, with outcome-weighted recall across five memory layers and 20 MCP tools. It records trades with SHA-256 chaining and RFC 3161 timestamp anchoring to support MiFID II and EU AI Act compliance, without executing trades itself.

## Use cases
- give my AI trading agent persistent memory across sessions
- recall past trades in similar market conditions before entering a position
- create a tamper-evident audit trail of AI trading decisions for regulators
- track drawdown and losing streaks to know when to stop trading
- log MT5 or forex expert advisor decisions automatically
- review strategy decay and behavioral drift in daily trading reflections

## When to choose
- you run AI agents (e.g., Claude) that trade and need cross-session memory of trades and outcomes
- you need regulator-ready, tamper-evident decision documentation for algorithmic trading
- you want pre-flight recall of similar past trades before opening positions
- you use MCP-compatible AI platforms and want a pip-installable memory layer

## When to avoid
- you need order execution or broker integration - it deliberately does not trade
- you expect new features or a hosted service - the project is in maintenance mode
- you need a general-purpose agent memory system unrelated to trading
- you require real-time low-latency trade execution support rather than post-trade logging

## Facets
- artifact type: service
- maturity: maintenance
- function: mcp, logging, monitoring, security
- domain: fintech, large-language-models, legal, developer-tools
- platform: python, self-hosted, cross-platform
- tags: mcp-server, trading-journal, tamper-evident, outcome-weighted-memory, miifid-ii, eu-ai-act, sha-256-chain, rfc-3161, forex, crypto, mt5, claude, audit-trail, memory, trading, ai-agents

## Member repositories
- mnemox-ai/tradememory-protocol (main) score 80

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:39.160571+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:38:21.001865+00:00, confidence not recorded.
  - readme: https://github.com/mnemox-ai/tradememory-protocol (fetched 2026-08-28T04:04:39.160571+00:00, sha 2d2737412715)
  - homepage: https://mnemox.ai/tradememory/ (fetched 2026-08-29T11:51:36.877737+00:00, sha 43f07a9db391)
  - registry_pypi: https://pypi.org/pypi/tradememory-protocol/json (fetched 2026-08-29T11:51:36.888786+00:00, sha c55cecd8d24a)
  - site_page: https://www.mnemox.ai/pricing (fetched 2026-08-29T11:51:36.886820+00:00, sha d8ccd66f6330)
- Data as of 2026-08-30T08:39:29.467469+00:00.