Experimental Project | This is exploratory, not production-ready. Read DISCLAIMER.md and DYOR.md before using. No warranties. No professional advice. Use at your own risk.
How brands should remember, forget, and learn from their own history.
When a brand agent needs to know "how did we handle this last time?", there's no system to answer that. Brand knowledge lives in people's heads and buried Slack threads. This repo builds the institutional brain for agentic organizations.
Created by Dragoon
git clone https://github.com/Dragoon0x/brand-memory-architecture.git
cd brand-memory-architecture
pip install -r requirements.txt
python storage/memory-store.py --init
python agents/memory-writer.py --event "shipping_crisis_jan_2025" --type crisis_response
python agents/memory-reader.py --situation "shipping_delay" --severity "high"- Crisis memories: How we handled problems
- Pricing memories: Pricing decisions and outcomes
- Campaign memories: Marketing campaign learnings
- Product memories: Product decisions and results
- Customer pattern memories: Recurring customer behaviors
A brand that remembers everything develops bias. The forgetting policy defines what the brand intentionally deprioritizes over time. Old failures that have been addressed. Outdated competitive intelligence. Strategic forgetting is as important as strategic remembering.
This project is experimental. See DISCLAIMER.md, DYOR.md, and LEGAL.md for important information.
MIT License. See LICENSE for full terms.
Part of the Agentic Brand Infrastructure collection by Dragoon