Browse categories
Explore
Fiverr Pro
English
$
USD
I will fix setup multi ai agent jarvis obsedian memory claude code hermes webui vps mac
Mustello
About this gig
Did you know most people who try to set up OpenClaw or Hermes Agent spend two to three days fighting errors before they get anything working?
The tutorials make it look easy. Then you run the install command and nothing happens. Or it runs but WhatsApp won't pair. Or it pairs once and drops every 40 minutes. Or you migrate to Hermes and quietly lose your entire agent personality, plugins, and Slack tokens with no error message telling you why. I fix all of that and set things up properly so they don't break again.
What I deliver:
- Jarvis AI multi-agent setup with role-based agent configuration
- Obsidian second brain: vault structure, tags, templates built for AI retrieval
- RAG pipeline grounding agents in your actual notes and documents
- Hermes Agent or OpenClaw with Paperclip persistent memory
- MCP server + Claude Code integration for full vault read/write access
- n8n automation: note capture from WhatsApp, Telegram, or web clips
- Agent activity dashboard showing run logs, outputs, and status
- Ollama + Kimi K2.5 for zero API cost local model option
- VPS, Mac Mini, Docker, or WSL2 deployment
- Fix for broken installs, LiveSync errors, plugin conflicts
Message me before ordering !!!
Get to know Mustello
Mustello
Fix and Deploy Hermes OpenClaw Claude Code n8n MCP Obsidian Expert
- FromNigeria
- Member sinceJun 2026
- Avg. response time1 hour
- Last delivery1 month
Languages
English, Spanish, French, German, Italian
Hi, I'm AI Automation Engineer, problem solver who helps businesses save time, reduce manual work, and scale with powerful AI solutions.
I specialize in OpenClaw, Claude Code, Hermes AI, n8n, WhatsApp Business API, AI agents, VPS deployment, cloud hosting, Google Play Console testing. I don't just deliver services. I solve complex technical challenges with reliable solutions.
Every project is handled with clear communication, attention to detail, and a commitment to quality. My goal is to deliver results that help your business grow & build long term relationships through exceptional service.
My Portfolio
FAQ
Do you build fully automated bots that trade without me watching?
Yes — all bots I build run autonomously on your VPS 24/7. They receive signals from TradingView, on-chain data, or prediction market APIs, process them through OpenClaw and Claude Code, and execute trades automatically. You get Telegram notifications for every trade so you're always informed
Can you add risk management logic to the bot?
Yes — risk management is included in the Standard and Premium packages. I configure position sizing limits, maximum drawdown stops, daily loss limits, and cooldown periods after consecutive losses. Claude Code handles the reasoning layer so the bot can adapt its risk parameters based on market cond
What happens if the bot breaks after delivery?
I include 3 days of post-delivery support on all packages. If the bot breaks due to RPC endpoint changes, API updates, or exchange maintenance within that window, I fix it at no charge. After the support window, I offer paid maintenance
Can you fix a broken OpenClaw or Hermes Agent install without starting over?
Yes. I run openclaw doctor repair or hermes doctor first, trace the root cause Node.js version, npm permissions, Docker, WSL2, firewall and recover your setup without a full wipe in most cases.
Can you migrate my OpenClaw memories, skills and API keys to Hermes Agent?
Yes. I run hermes claw migrate dry run first so you preview exactly what transfers, then resolve conflicts and verify memories, skills, and API keys all loaded correctly before marking the job done.
How do you fix Ollama not connecting to n8n inside WSL2?
The issue is Ollama running on the Windows host while n8n runs inside WSL2 localhost does not bridge between them. I install Ollama inside WSL2 or use Docker Compose to put both on the same network, then test the n8n Ollama Chat Model node
What models do you set up for local or OpenRouter AI agents?
For local inference via Ollama Llama 4 Maverick, Qwen 3, Mistral, DeepSeek, Hermes 4. For OpenRouter DeepSeek V4 for budget, Claude Sonnet for quality tool calling, Qwen 3.6 free tier for zero cost.
Do you set up multi-agent routing, sub-agents and MCP servers too?
Yes. Multi-agent routing gives each agent its own channel, SOUL.md persona, and tool permissions. Sub-agents run background tasks and post results back to the main agent. MCP servers extend tool capabilities beyond built-ins. I set up all three and test each before delivery.
