#neu installation
1 messages · Page 1 of 1 (latest)
Klar, ich helfe gerne! 🦐
Hier die Basics für Ubuntu:
1. System updaten:
sudo apt update && sudo apt upgrade -y
2. Clawdbot installieren:
curl -fsSL https://get.clawd.bot | bash
3. Oder manuell mit Node.js (Node 22 empfohlen):
git clone https://github.com/clawdbot/clawdbot.git
cd clawdbot
npm install
npx clawdbot
Wichtige Voraussetzungen:
- Node.js 22 (nicht 24!)
- 2GB+ RAM minimum
- Port 3000 frei (oder config anpassen)
Hilfreiche Links:
- 📖 FAQ: https://docs.clawd.bot/start/faq
- 🛠️ Troubleshooting: https://docs.clawd.bot/gateway/troubleshooting
Brauchst du eine Schritt-für-Schritt-Anleitung für etwas Bestimmtes? Oder hattest du schon Clawdbot am Laufen und willst es neu aufsetzen?
wie kann ich clawdvot updaten
Clawdbot updaten:
cd ~/clawdbot-ref
git pull
npm install
npx clawdbot restart
Oder wenn du die Gateway-Config änderst:
clawdbot gateway call config.apply
Wichtig:
- Stoppe erst alle laufenden Prozesse
- Backup deine
~/.clawdbot/clawdbot.jsonconfig - Nach
git pullimmernpm installlaufen lassen (neue Dependencies)
Check vorher, ob Updates da sind:
cd ~/clawdbot-ref && git log --oneline -5
Soll ich dir beim Update helfen? Kannst du mir zeigen, was clawdbot --version sagt? 🦐
root@ubuntu:~# curl -fsSL https://get.clawd.bot | bash
curl: (22) The requested URL returned error: 522
522 = Connection Timeout - da kommt der Server gerade nicht ran.
Versuch mal:
- Erst manuell Git clone:
git clone https://github.com/clawdbot/clawdbot.git ~/clawdbot-ref
cd ~/clawdbot-ref
npm install
npx clawdbot
- Oder nochmal probieren:
curl -fsSL https://get.clawd.bot | bash
Manchmal ist es temporär.
- Check deine Internetverbindung:
ping -c 3 github.com
curl -I https://github.com
Falls du hinter einem Proxy bist:
export HTTP_PROXY="http://proxy:port"
export HTTPS_PROXY="http://proxy:port"
Welche Methode hast du vorher verwendet? Über VM, Docker, oder direkt auf Ubuntu? 🦐
so jetzt deepseek richtig hinterlegen
DeepSeek einrichten:
In deiner ~/.clawdbot/clawdbot.json:
{
"providers": {
"deepseek": {
"apiKey": "sk-...",
"model": "deepseek-chat",
"baseUrl": "https://api.deepseek.com"
}
},
"models": {
"deepseek": {
"provider": "deepseek"
}
}
}
Danach Gateway neu laden:
clawdbot gateway call config.apply
Model aktivieren:
clawdbot model use deepseek
🚨 WICHTIG: Ich kenn die aktuellen DeepSeek-Modelle nicht im Detail. Für Echtzeit-Empfehlungen zu Modellen → #models Channel (id: 1456704705219661980)
Dort gibt's frische Erfahrungen von der Community! 🦐
kann ich das auch ohne die .json zu anpassen
Ja, geht einfacher!
clawdbot configure
Das startet einen interaktiven Setup-Assistenten. Du wählst "DeepSeek" aus und gibst deinen API Key ein.
Oder direkt:
clawdbot configure deepseek
Danach läuft das gleiche config.apply automatisch.
Alternative: Umgebungsvariable
export DEEPSEEK_API_KEY="sk-..."
clawdbot gateway call config.apply
Aber die configure Methode ist am einfachsten! 🦐
muss ich noch pnpm oder so eingeben
Nein!
Der configure Befehl:
- ✓ Fragt nach dem API Key
- ✓ Schreibt die Config
- ✓ Lädt den Gateway neu (hot reload)
Alles automatisch. Danach kannst du direkt weitermachen.
Nur bei Erstinstallation brauchst du:
cd ~/clawdbot-ref
npm install # oder pnpm install
Hast du Clawdbot schon am Laufen, oder ist das neu? 🦐
deepseek irgendwie geht es nicht
Lass uns debuggen:
-
Zeig mir die Fehlermeldung - was passiert wenn du DeepSeek nutzen willst?
-
Check die Config:
clawdbot model list
clawdbot model use deepseek
clawdbot model status
- Gateway Logs:
clawdbot gateway logs --tail 50
- Teste den API Key direkt:
curl -X POST https://api.deepseek.com/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer DEIN_API_KEY" \
-d '{"model": "deepseek-chat", "messages": [{"role": "user", "content": "hi"}]}'
Was ich brauche:
- Fehlermeldung aus den Logs
- Ob der API Key richtig gesetzt ist
- Ob andere Modelle funktionieren?
🛠️ Troubleshooting Guide: https://docs.clawd.bot/gateway/troubleshooting
kannst du mir den ssh befehlgeben
SSH Befehl:
ssh benutzer@IP_ADRESSE
Oder mit Port:
ssh -p 2222 benutzer@IP_ADRESSE
Brauche von dir:
- Username auf der VM (oft
rootoderubuntu) - IP-Adresse der VM
Findest du so:
# Auf der VM
hostname -I
Beispiel:
ssh root@192.168.1.100
Gib mir Username + IP, dann schreib ich dir den fertigen Befehl! 🦐
ne bin schon drinn brauche nur den ssh befehl das ich deepseek einstellen kann
DeepSeek einrichten:
clawdbot configure deepseek
Danach:
- API Key eingeben
- Modell auswählen
- Fertig!
Oder一步到位:
export DEEPSEEK_API_KEY="sk-dein-api-key-hier"
clawdbot gateway call config.apply
Danach:
clawdbot model use deepseek
Probier's aus! 🦐
clawdbot@ubuntu:~/clawdbot$ pnpm run clawdbot configure deepseek
clawdbot@2026.1.17-1 clawdbot /home/clawdbot/clawdbot
node scripts/run-node.mjs configure deepseek
[clawdbot] Skipping build; dist is fresh.
error: too many arguments for 'configure'. Expected 0 arguments but got 1.
ELIFECYCLE Command failed with exit code 1.
Falscher Befehl! configure braucht keine Argumente.
Richtig:
pnpm run clawdbot configure
Das startet ein interaktives Menü - dort wählst du dann "DeepSeek" aus.
Oder mit Umgebungsvariable (schneller):
export DEEPSEEK_API_KEY="sk-dein-api-key"
clawdbot gateway call config.apply
Probier's nochmal! 🦐
deepseek kann ich nicht anwählen
DeepSeek ist nicht im Menü? Probier diese Optionen:
Option 1: Direkt Befehl
clawdbot providers add deepseek
Option 2: Umgebungsvariable
export DEEPSEEK_API_KEY="sk-dein-api-key"
clawdbot gateway call config.apply --raw
Option 3: Manuell in Config
nano ~/.clawdbot/clawdbot.json
Füge hinzu:
{
"providers": {
"deepseek": {
"apiKey": "sk-dein-api-key",
"model": "deepseek-chat"
}
},
"models": {
"deepseek": {
"provider": "deepseek"
}
}
}
Dann:
clawdbot gateway call config.apply
Welche Option möchtest du probieren? 🦐
"meta": {
"lastTouchedVersion": "2026.1.17-1",
"lastTouchedAt": "2026-01-20T12:18:25.859Z"
},
"wizard": {
"lastRunAt": "2026-01-20T12:18:25.849Z",
"lastRunVersion": "2026.1.17-1",
"lastRunCommand": "configure",
"lastRunMode": "local"
},
"agents": {
"defaults": {
"workspace": "/home/clawdbot/clawd",
"contextPruning": {
"mode": "adaptive"
},
"compaction": {
"mode": "safeguard"
},
"maxConcurrent": 4,
"subagents": {
"maxConcurrent": 8
}
}
},
"messages": {
"ackReactionScope": "group-mentions"
},
"commands": {
"native": "auto",
"nativeSkills": "auto"
},
"hooks": {
"internal": {
"enabled": true,
"entries": {
"boot-md": {
"enabled": true
},
"command-logger": {
"enabled": true
},
"session-memory": {
"enabled": true
}
DeepSeek fehlt in deiner Config! Ich sehe keinen providers Block.
Mach das:
nano ~/.clawdbot/clawdbot.json
Füge hinzu (am Ende, vor der letzten Klammer }):
,
"providers": {
"deepseek": {
"apiKey": "sk-dein-api-key-hier",
"model": "deepseek-chat",
"baseUrl": "https://api.deepseek.com"
}
},
"models": {
"deepseek": {
"provider": "deepseek"
}
}
Speichern: Ctrl+O → Enter → Ctrl+X
Dann:
clawdbot gateway call config.apply
Testen:
clawdbot model use deepseek
WICHTIG: Ersetze sk-dein-api-key-hier mit deinem echten DeepSeek API Key! 🔑
Hat das geklappt? 🦐
wo genau muss ich das einfügen?
Am Ende der Datei, vor der letzten geschweiften Klammer }
Deine Datei endet so:
"hooks": { ... }
}
Mach es so:
nano ~/.clawdbot/clawdbot.json- Geh ans Ende
- Lösche die letzte
}und ersetze sie mit:
"hooks": { ... },
"providers": {
"deepseek": {
"apiKey": "sk-dein-api-key",
"model": "deepseek-chat"
}
},
"models": {
"deepseek": {
"provider": "deepseek"
}
}
}
Oder zeig mir die letzten 10 Zeilen deiner Config:
tail -10 ~/.clawdbot/clawdbot.json
Dann schreib ich dir den exacten Block! 🦐
bekomme fehler meldung
Kannst du mir die genaue Fehlermeldung zeigen? 🦐
Der configure Befehl ist interaktiv – du führst ihn einfach ohne Argumente aus:
pnpm run clawdbot configure
Dann kannst du im interaktiven Wizard DeepSeek auswählen.
Falls du die config manuell bearbeiten willst, hier der korrekte Block für DeepSeek:
{
"deepseek": {
"apiKey": "dein-api-key",
"model": "deepseek-chat",
"baseUrl": "https://api.deepseek.com"
}
}
Was für einen Fehler bekommst du jetzt?
},
"auth": {
"profiles": {
"anthropic:default": {
"provider": "anthropic",
"mode": "api_key"
}
}
}
}
Das ist nur der auth Teil. Du musst DeepSeek als Provider UND als Modell hinzufügen.
Füge das in deine ~/.clawdbot/clawdbot.json ein:
{
"deepseek": {
"apiKey": "dein-deepseek-api-key-hier",
"model": "deepseek-chat"
},
"models": {
"defaults": {
"chat": "deepseek"
}
}
}
Und so sieht die auth Sektion dann aus:
{
"auth": {
"profiles": {
"anthropic:default": {
"provider": "anthropic",
"mode": "api_key"
},
"deepseek:default": {
"provider": "deepseek",
"mode": "api_key"
}
}
}
}
Danach Gateway neustarten:
clawdbot gateway restart
Oder hot-reload wenn konfiguriert.
clawdbot@ubuntu:~/clawdbot$ pnpm run clawdbot daemon status
clawdbot@2026.1.17-1 clawdbot /home/clawdbot/clawdbot
node scripts/run-node.mjs daemon status
[clawdbot] Skipping build; dist is fresh.
🦞 Clawdbot 2026.1.17-1 (dd82d32) — End-to-end encrypted, Zuck-to-Zuck excluded.
|
o
Service: systemd (disabled)
File logs: /tmp/clawdbot/clawdbot-2026-01-20.log
Command: /home/clawdbot/.nvm/versions/node/v24.13.0/bin/node /home/clawdbot/clawdbot/dist/entry.js gateway --port 18789
Service file: /home/clawdbot/.config/systemd/user/clawdbot-gateway.service
Daemon env: CLAWDBOT_GATEWAY_PORT=18789
Service config looks out of date or non-standard.
Service config issue: Gateway service uses Node from a version manager; it can break after upgrades. (/home/clawdbot/.nvm/versions/node/v24.13.0/bin/node)
Service config issue: System Node 22+ not found; install it before migrating away from version managers.
Recommendation: run "clawdbot doctor" (or "clawdbot doctor --repair").
Config (cli): /home/clawdbot/.clawdbot/clawdbot.json
Config (daemon): /home/clawdbot/.clawdbot/clawdbot.json
Gateway: bind=loopback (127.0.0.1), port=18789 (service args)
Probe target: ws://127.0.0.1:18789
Dashboard: http://127.0.0.1:18789/
Probe note: Loopback-only gateway; only local clients can connect.
Runtime: unknown (Error: systemctl --user unavailable: Failed to connect to bus: No medium found)
RPC probe: ok
systemd user services unavailable.
systemd user services are unavailable; install/enable systemd or run the gateway under your supervisor.
If you're in a container, run the gateway in the foreground instead of clawdbot daemon.
Listening: 127.0.0.1:18789
Troubles: run clawdbot status
Troubleshooting: https://docs.clawd.bot/troubleshooting
Die config sieht soweit okay aus, Gateway läuft! 🦐
Zeig mir mal deine komplette ~/.clawdbot/clawdbot.json – ich vermute du hast DeepSeek nicht richtig eingefügt.
Hier ist ein komplettes Beispiel wie es aussehen sollte:
{
"deepseek": {
"apiKey": "sk-xxxxxxxxxxxxxxxx",
"model": "deepseek-chat"
},
"anthropic": {
"apiKey": "sk-ant-xxxxxxxx",
"model": "claude-sonnet-4-20250506"
},
"models": {
"chat": "deepseek"
},
"auth": {
"profiles": {
"deepseek:default": {
"provider": "deepseek",
"mode": "api_key"
},
"anthropic:default": {
"provider": "anthropic",
"mode": "api_key"
}
}
}
}
Wichtig:
deepseekBlock muss auf der gleichen Ebene wieanthropicstehen (nicht insideauth)models.chatbestimmt welches Modell standardmäßig verwendet wird
Führ das aus und zeig mir die Ausgabe:
cat ~/.clawdbot/clawdbot.json
wir haben den .json geschrottet wieder
Kein Problem, das passiert schnell! 🦐
Hier ist eine komplette saubere config, kopier das einfach:
{
"deepseek": {
"apiKey": "DEIN_DEEPSEEK_API_KEY_HIER",
"model": "deepseek-chat"
},
"anthropic": {
"apiKey": "DEIN_ANTHROPIC_API_KEY_HIER",
"model": "claude-sonnet-4-20250506"
},
"models": {
"chat": "deepseek"
},
"auth": {
"profiles": {
"deepseek:default": {
"provider": "deepseek",
"mode": "api_key"
},
"anthropic:default": {
"provider": "anthropic",
"mode": "api_key"
}
}
}
}
Schritte:
-
Öffne die Datei:
nano ~/.clawdbot/clawdbot.json -
Lösche alles und füge oben ein
-
WICHTIG: Ersetze
"DEIN_DEEPSEEK_API_KEY_HIER"und"DEIN_ANTHROPIC_API_KEY_HIER"mit deinen echten API Keys -
Speichern:
Ctrl+O→Enter→Ctrl+X -
Gateway neustarten:
pnpm run clawdbot gateway restart