Skip to content

ariaxhan/memory-pool

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

44 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Synthesis Pool

Memory archaeology system that treats life experiences as matter to be processed, synthesized, and analyzed.

Overview

Synthesis Pool provides a metaphor-driven interface for personal data processing:

  • Memory Matter: Raw inputs (journals, memories, artifacts)
  • Pool: Processing substrate that metabolizes inputs
  • Synthesis Outputs: Distilled insights (medicine/material/fuel/toxin)
  • Pool States: Health snapshots over time

Quick Start

# Clone
git clone https://github.com/ariaxhan/memory-pool.git
cd memory-pool

# Backend setup
python -m venv venv && source venv/bin/activate
pip install -r requirements.txt
cd api && pip install -r requirements.txt && cd ..

# Frontend setup
cd frontend && npm install && cd ..

# Initialize data
mkdir -p data && cp data.example/* data/

# Run (two terminals)
cd api && python -m uvicorn main:app --reload --port 8000
cd frontend && npm run dev

Access:

Documentation

Document Description
docs/SETUP.md Full installation and troubleshooting
docs/ARCHITECTURE.md System design and data flow
docs/SCHEMA.md CSV data structures
api/README.md API reference

Stack

  • Backend: FastAPI + Python 3.11+
  • Frontend: SvelteKit 2.49 + Svelte 5 + TypeScript
  • Data: CSV files (local, portable, human-readable)
  • CLI: Click

CLI Usage

source venv/bin/activate

# Feed memory
python pool.py feed --date 2024-12-15 --source journal \
  --raw "Content" --layers work|self --toxicity 3

# Observe pool state
python pool.py observe

# Query memories
python pool.py query --layer work --after 2024-01-01

# Harvest synthesis
python pool.py harvest --type medicine --from mm_001,mm_002 \
  --concentration 4 --label "Insight Name"

Philosophy

  • CSV-Native: Simple, portable, human-readable storage
  • Local-First: No cloud dependencies
  • Privacy-Focused: Your data stays on your machine
  • Metaphor-Driven: Intuitive language for data processing

Contributing

Contributions welcome. Keep it privacy-focused and maintain the metaphor-driven design.

License

MIT - See LICENSE