YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Hermes Memory Synchronization System

This repository contains tools for backing up and restoring Hermes AI agent state to Hugging Face Datasets.

Files

  1. hermes_sync.py - The main synchronization script for backing up and restoring Hermes state
  2. AGENTS.md - Documentation for the Hermes Memory Synchronization System

hermes_sync.py

This script provides a complete backup and restore solution for Hermes AI agent state. It can:

  • Create backups of all Hermes state files
  • Upload backups to Hugging Face Datasets
  • Restore from backups stored on Hugging Face
  • Run automatic hourly backups
  • Automatically clean up local backup files after upload

Usage

Download the script

wget https://huggingface.co/datasets/R1000/Hermes-Memory/resolve/main/hermes_sync.py

Set your Hugging Face token

export HF_TOKEN=your_huggingface_token_here

Create and upload a backup

python3 hermes_sync.py backup --upload

Restore from the latest backup

python3 hermes_sync.py restore

List available backups

python3 hermes_sync.py list

Run automatic backup (for cron jobs)

python3 hermes_sync.py auto-backup

What gets backed up

The script backs up all essential Hermes state:

  • state.db + WAL files - core KV state
  • response_store.db - chat response cache
  • sessions/ - session transcripts
  • skills/ - user-installed skills
  • cron/ - cron job definitions
  • memories/ - persistent memories
  • auth.json - OAuth tokens
  • channel_directory.json - registered channels
  • config.yaml - active configuration
  • gateway_state.json - gateway routing state
  • .env - environment overrides
  • SOUL.md - persona
  • .skills_prompt_snapshot.json - skill snapshot

Security

  • Token is stored only in environment variables
  • Local backup files are automatically deleted after successful upload
  • No tokens are stored in the repository
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support