I will setup jarvis ai assistant hermes agent multi agent system obsidian rag ai agent


1 order in queue
About this gig
Want your own Jarvis-style AI agent running 24/7, privately and securely? I will set up OpenClaw (formerly Clawdbot/Moltbot) or Hermes Agent on your VPS, Mac mini or PC and connect it to Telegram or WhatsApp.
Many self-hosted AI agents end up exposed online with open dashboards and weak tokens. I set yours up the secure way.
What you get:
- Clean install on the latest stable release
- Private access (loopback + Tailscale) and a strong gateway token
- Telegram or WhatsApp with pairing, so only you can command it
- Claude, GPT, Gemini, DeepSeek or local Ollama models, with fallbacks
- Security audit, with every finding fixed before delivery
- Custom skills, MCP servers and n8n automations (higher packages)
- A clear runbook: restart, update and recovery steps
Works on: Ubuntu VPS, Docker, macOS, Windows/WSL2 and Raspberry Pi.
Not sure whether OpenClaw or Hermes Agent fits you? Message me before ordering and I'll recommend the right one for your use case.
Need extra channels, a team setup or a custom integration? Send me a message for a custom offer.
Get to know Joshua E
AI Agentic Developer, Multi Agent System, Jarvis AI, Relevance AI
- FromUnited Kingdom
- Member sinceSep 2026
- Avg. response time1 hour
Languages
English, Spanish, French, German, Italian
FAQ
OpenClaw or Hermes Agent: which should I choose?
OpenClaw is best for many chat channels, phone and desktop control, and fine-grained settings. Hermes Agent is best if you want an agent that learns your routines over time. Tell me what you need and I'll recommend one, or set up both.
Do I need a VPS or a powerful computer?
No. It runs on a Mac, Windows or Linux PC with no GPU needed. For a 24/7 assistant, a small VPS ($5–15/month) is best, and I'll help you choose one. A GPU only matters if you want local AI models.
How much will it cost me each month?
Typically $5–15 for hosting plus your AI model usage, which you pay directly to the provider. Light use can cost a few dollars a month; local models cost nothing per message.
Which AI models can I use?
Claude, GPT, Gemini, DeepSeek, Qwen, Kimi, OpenRouter, or free local models through Ollama or LM Studio. I also set a backup model so your agent keeps working if one provider fails.
Do you need my passwords or API keys?
No passwords. You create your own accounts and keys. I work over screen share or temporary server access, which you remove after delivery. Your keys stay on your machine.
Is it safe to run an AI agent with access to my computer?
Yes, when set up properly. I keep it off the public internet, enable pairing so strangers can't use it, limit what it can run, and run the built-in security audit before delivery.
My OpenClaw or Clawdbot install is broken. Can you fix it?
Yes. Order the Basic package for one issue, or message me the error first. I fix gateway, WhatsApp, Telegram, Ollama, Docker and update problems, and I migrate old Clawdbot/Moltbot setups.
Can I use WhatsApp with my own number?
You can, but I recommend a separate number. WhatsApp support uses an unofficial client, so a dedicated number keeps your personal chats separate and avoids risk to your main account.
What support do I get after delivery?
Each package includes support days for questions and fixes. You also get a runbook for restarts, updates and recovery. Monthly maintenance is available if you want me to handle updates.
What do I need to provide before you start?
Your goal for the agent, your preferred chat app, an AI model account (or a local model), and a VPS or computer to install on. If you don't have these yet, I'll guide you through getting them.

