ASSISTANT (hermes, glm-5.1)
Before nuking, let's catalog what you have so nothing important is lost. Let me scan the key stuff:
2026-07-05T00:29
ASSISTANT (hermes, glm-5.1)
Here's your complete rebuild inventory:
---
**NUKE & REBUILD CHECKLIST**
**1. TRADING BOTS (3 active)**
| Bot | Dir | Service | Entry Point |
|---|---|---|---|
| claude_bot | /home/ubuntu/bots/Claude_bot/ (540MB!) | claude_bot.service | unified_strategy.py |
| lt_3candle | /home/ubuntu/bots/three_candle_bot/ | lt_3candle_bot.service | lt_3candle_bot.py |
| morning_sniper | /home/ubuntu/bots/morning_sniper/ | morning_sniper_bot.service | sniper_bot.py |
Config files: `bots/*/config*.py` — broker credentials live here
DBs: 11 .db files across bots (trading history, candles)
**2. OPENALGO (3 instances behind nginx)**
- Flattrade: `/var/python/openalgo-flask/flattrade-openalgo.theworkpc.com-flattrade/`
- Shoonya: `/var/python/openalgo-flask/shoonya-openalgo.theworkpc.com-shoonya/`
- Kotak: `/var/python/openalgo-flask/kotak-openalgo.theworkpc.com-kotak/`
- Nginx sites: 4 configs (chart, flattrade, kotak, shoonya)
- 2 token bridges running as systemd services
- Domain: `theworkpc.com` with subdomains
**3. SCRIPTS** (`/home/ubuntu/scripts/`)
Key ones: `auto_db_sync.py`, `post_login_flush.sh`, `token_bridge.py`, `market_ready.sh`, `flattrade_master_contract_refresh.py`, `sync_hermes_to_gemini.py`, `sync_claude_to_gemini.py`
**4. CRONTAB** — 13 entries (health monitors, EOD scripts, DB syncs, master contract refresh, gemini syncs)
**5. HERMES** — 3.9GB, config.yaml + .env with API keys, skills, cron jobs, memories
**6. OTHER**
- `.bashrc` — custom aliases (claude-mem, claude-pro, claude-mini, morning_fix), PATH entries (bun, opencode, bots dir)
- Claude Code + Gemini + Copilot configs
- Telegram bot (claude_tg_bot.service)
- Python 3.13, Node 22, nginx 1.26, gunicorn
---
**WHAT TO BACK UP BEFORE NUKING:**
```
1. /home/ubuntu/bots/ — all bot code + DBs + configs (560MB)
2. /home/ubuntu/scripts/ — utility scripts
3. /home/ubuntu/.hermes/ — config, skills, memories, auth.json, .env
4. /home/ubuntu/.bashrc — aliases + env vars
5. /home/ubuntu/.claude/ — claude code config
6. /var/python/openalgo-flask/ — all 3 OpenAlgo instances (4.9GB)
7. /etc/nginx/sites-available/ — nginx configs
8. /etc/systemd/system/*bot* — all bot service files
9. /etc/systemd/system/*openalgo* + *token-bridge*
10. crontab -l — save the crontab
```
**SUGGESTION:** Tar up the critical stuff and download it, then rebuild clean. Want me to:
1. Create a backup script that tars everything above into a single archive?
2. Set up an S3/rclone upload so you can pull it after rebuild?
3. Just tar it locally and you scp it down?
2026-07-05T00:29
ASSISTANT (hermes, glm-5.1)
The tar files didn't get created properly. Let me debug this:
2026-07-05T00:29