Skip to content

README.md

Latest commit

 

History

History
90 lines (63 loc) · 2.29 KB

README.md

File metadata and controls

90 lines (63 loc) · 2.29 KB
Raw

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