Menu
Le prompt qui vérifie d'abord la source est sélectionné par défaut. Vous pouvez passer à une commande directe ou télécharger une copie locale.
Lisez SKILL.md et les fichiers associés affichés par SkillsMP avant de décider de l'installer.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Une commande directe contourne le prompt de vérification. Examinez la source avant de l'exécuter.
npx skills add https://github.com/Turnipnator/Binance_Bot --skill healthcheckLa commande reste sur une seule ligne. Faites défiler horizontalement pour la vérifier avant de la copier.
Vous préférez une copie locale ? Téléchargez les fichiers actuellement disponibles dans SkillsMP.
Affichage de SKILL.md
| name | healthcheck |
| description | Run a comprehensive health check on the Binance trading bot |
Run a comprehensive health check on the binance-trading-bot. Work through each section systematically and provide a summary dashboard at the end.
Read VPS connection details from CLAUDE.local.md (contains server IP, SSH key path, bot path).
docker psssh -i <SSH_KEY> <USER>@<VPS_IP> "docker ps --format '{{.Names}}\t{{.Status}}\t{{.RunningFor}}' | grep binance"
ssh -i <SSH_KEY> <USER>@<VPS_IP> "docker logs binance-trading-bot --tail 100 2>&1"
ssh -i <SSH_KEY> <USER>@<VPS_IP> "docker logs binance-trading-bot 2>&1 | grep -iE 'error|warn|fail|disconnect|reconnect|rate.limit' | tail -20"
ssh -i <SSH_KEY> <USER>@<VPS_IP> "ls -la <BOT_PATH>/data/"
ssh -i <SSH_KEY> <USER>@<VPS_IP> "docker exec binance-trading-bot cat /app/data/state.json 2>/dev/null || echo 'No state file'"
ssh -i <SSH_KEY> <USER>@<VPS_IP> "docker exec binance-trading-bot cat /app/data/trading_stats.json 2>/dev/null || echo 'No stats file'"
ssh -i <SSH_KEY> <USER>@<VPS_IP> "docker exec binance-trading-bot cat /app/data/positions.json 2>/dev/null || echo 'No positions file'"
ssh -i <SSH_KEY> <USER>@<VPS_IP> "free -h && echo '---' && df -h / && echo '---' && top -bn1 | head -12"
ssh -i <SSH_KEY> <USER>@<VPS_IP> "grep -E 'ENABLE_|TRADING_PAIRS|MODE|STRATEGY' <BOT_PATH>/.env 2>/dev/null | head -15"
The mean-reversion strategy went live 2026-07-02 with an unproven live edge (deployed on a 53-trade backtest, PF 1.82 @0.2% fee). This section tracks whether the real fills hold up. Run the monitor (stdlib-only, runs on the VPS host — no container/rebuild needed):
ssh -i <SSH_KEY> <USER>@<VPS_IP> "python3 /opt/Binance_Bot/mr_monitor.py"
Interpret the output:
ENABLE_MEAN_REVERSION=false).stop_loss exits (backtest exits mostly at the EMA20 target) → dips reverting less than modeled.strategy in trades.json.ssh -i <SSH_KEY> <USER>@<VPS_IP> "docker logs binance-trading-bot 2>&1 | grep -c 'MR pre-gate VETO'"
Count these across checks (log retention ~18h, so note the count each run). A recurring nonzero count = the pre-gate is suppressing backtest-valid entries; factor it into any live-vs-backtest verdict alongside the tick-exit mismatch.
The monitor also prints a STRATEGY HEAD-TO-HEAD: momentum vs mean_reversion, both under current rules (trades since 2026-07-02 only — earlier momentum trades are old-rule and NOT comparable). It reports each strategy's n / WR / net$ / PF / expectancy $ per trade. Read BOTH: net$ = what grows the account, expectancy = edge quality normalised for the fact MR trades far more often than momentum. Needs ~15 trades each before it's meaningful.
Report MR's live line, the backtest line, the head-to-head table, and a one-line verdict (on track / slipping / disable; which strategy leads).
Provide prioritised recommendations:
Present a quick status summary table:
| Check | Status | Notes |
|---|---|---|
| Process Running | ?/? | |
| Logs Healthy | ?/?/? | |
| Signals Active | ?/? | |
| Resources OK | ?/?/? | |
| WebSocket Connected | ?/? | |
| Strategy Edge (momentum) | ?/?/? | |
| Mean-Reversion vs Backtest | ?/?/? | n trades; on track / slipping / dormant |
Traffic light summary: ? All good / ? Minor issues / ? Needs attention