Key Takeaways

• Flexibility is the most complete computer-control AI I tested — it built a working website in roughly 30 seconds and analyzed Nvidia stock data in seconds
• Versip (versip.com), now owned by Anthropic, offers solid browser automation but requires a paid upgrade for full computer control
• Claude Code and similar tools can now control computers through “cowork” systems, though pricing varies by region
• A Brazilian no-code AI platform hit 3 million users and generated approximately $3 million in just 48 hours after launch
• The real money isn’t in using these tools — it’s in building automated systems you can sell or run 24/7

Flexibility: The Standout Performer

I started with Flexibility because it’s the one I actually use daily. They recently launched a computer control feature that lets the AI operate your browser and desktop directly. I tested two scenarios that matter for my business.

First, I asked it to analyze Nvidia’s 10-year stock trend. Within seconds, it navigated to Nvidia’s page, pulled the data, and generated a visual analysis — something that would take me 15-20 minutes manually. The speed genuinely surprised me. I watched it click through pages, extract information, and build the summary in real-time.

Then I pushed harder: “Build me a website.” From my phone, I gave the command. Flexibility opened the browser, created the index file, built supporting pages, and had a functional interactive site ready in approximately 30 seconds. For context, this same process used to take me 5 minutes even with other AI tools, and hours if done manually.

The key detail: Flexibility runs on Claude Sonnet 4.6 in its current language model. It can handle geographic design requests, multi-step workflows, and complex browser automation. I now use it to control my computers remotely via phone commands — starting tasks while I’m out and letting them run.

Versip and Claude Code: The Anthropic Alternative

Anthropic acquired Versip, and they’ve integrated similar computer control through Claude Code. The system works, but there’s a catch: you need to upgrade from the free plan to activate full browser and computer control. They use what’s called a “cowork” system — essentially a virtual workspace where the AI operates on your behalf.

I tested the workflow, and it’s capable. The pricing shows monthly tiers, though I couldn’t verify exact regional equivalents for Turkey versus UK rates. For my purposes, it works but feels slightly less integrated than Flexibility for rapid e-commerce tasks.

The Real Business Model: Build Systems, Not Just Use Tools

Here’s what I tell everyone who asks me which AI to buy: the tools are cheap; the systems are valuable.

I asked Flexibility directly: “What business models could I build with automation?” It suggested eBay dropshipping, Facebook dropshipping, and affiliate marketing — the three I consistently recommend because the earning limits are essentially uncapped if you execute correctly.

But the bigger opportunity? Building the automation itself and selling it.

I reference Empire Flippers and similar marketplaces constantly. I’ve seen SaaS projects sell for $5.5 million with $149,000 monthly revenue. I’ve watched YouTube channels list for £122,000. The pattern is clear: build an automated asset, then exit.

A Brazilian company just proved this at massive scale. They built a no-code AI platform, developed it with 500,000 paid users in roughly two weeks to a month, launched, and generated approximately $3 million in 48 hours — about 100 million Turkish Lira. I verified this story through multiple sources. Their servers crashed from traffic, which tells you everything about demand.

What I Actually Run: My Automation Stack

I’m transparent about my numbers because context matters. One of my systems generates approximately £20,000 monthly, with some months hitting near £400,000 in total flow. This isn’t profit — it’s revenue through automated e-commerce and affiliate systems I’ve built over years.

My eBay stores run with AI handling:

• Product research and niche identification
• Listing creation and optimization
• Price repricing against competitors
• Customer service and email responses
• Ad content generation
• Complete workflow management

I used to employ five people for my eBay operations. I don’t anymore. The AI works 24/7, doesn’t request time off, doesn’t get salary increases, and doesn’t have personal problems affecting output. It reports performance, and I give it better instructions.

But I need to be honest about something critical: these systems took time to build. I’ve seen people expect overnight millions. The Brazilian company’s $3 million in 48 hours? That came after intensive development. My own systems required months of refinement. Automation grows with your process — it’s not a lottery ticket.

The Mindset Shift Nobody Talks About

I’ve mentored hundreds of people through my programs. The pattern is predictable. Most watch, few act. Some buy every course but never implement. Others jump between projects before any gain traction.

The difference between people who build automated income and those who don’t? One group watches success stories; the other reverse-engineers them.

When I explain that someone built a $3 million system in 48 hours, most people react with disbelief or immediately ask for money. Very few ask: “How did they structure the offer? What was the user acquisition model? How can I adapt this?”

I’ve written about this extensively in my book — not as a success story, but as a failure analysis. I made expensive mistakes learning this. Time is the resource you can’t recover, and trying to self-teach complex automation without guidance costs more in lost time than any course or service.

FAQ

Which AI computer control tool is best for beginners?

Based on my testing, Flexibility has the lowest friction for newcomers. The phone-based command system means you can start automation without complex setup. Versip/Claude Code works well but requires navigating paid tiers for full functionality.

Can these tools really replace employees for e-commerce?

In my direct experience, yes — with caveats. I eliminated five eBay staff positions using AI automation. However, you need existing business knowledge. The AI handles execution, but you must direct strategy, handle edge cases, and maintain account health. It’s augmentation at small scale, replacement at scale.

How much does it cost to start with these AI tools?

Many core functions are free or low-cost. Flexibility and similar platforms offer substantial free tiers. My School community includes links to access premium AI models (including Claude variants) at reduced or no cost for students. The real investment isn’t software — it’s learning to build systems that leverage the software.

What’s the realistic timeline to automated income?

From my observation across hundreds of cases: minimum 6-12 months for a functioning automated system, and 1-2 years for significant income. The Brazilian $3 million case was exceptional and preceded by rapid development. Most people overestimate short-term results and underestimate compound growth from consistent automation.

Conclusion

The AI tools I tested in 2025 deliver on their core promise: genuine computer automation that works while you sleep. Flexibility and Versip/Claude Code are the leaders I’ve validated personally. But tools are just tools. The opportunity — and the challenge — is building systems so automated that they become sellable assets.

I run these systems daily. I see the revenue reports. I also see the months of work behind the “passive” income. If you’re considering this path, my advice is direct: pick one model, commit to 12 months of building, and use AI to accelerate every step. The bridge from employed to automated is crossing that commitment gap — and nobody can walk it for you.

Watch the full video (in Turkish — English subtitles available):

Tools & Community

• TurkoLister — the AI listing tool I use to turn Amazon products into optimized eBay UK listings in about 60 seconds (from £4.99/month, £1 one-week trial).
• AI & E-commerce Community — my Turkish-speaking community ($19/month) with weekly live sessions.
• Subscribe on YouTube — new experiments every week.

The post I Tested 5 AI Tools That Control Your Computer — Here’s What Actually Works appeared first on Digital Market Mentoring.

Key Takeaways

• I built a 24/7 AI agency using 5 tools for roughly $40/month total, down from £400+ I used to spend
• Paperclip acts as CEO with 167 ready-made agent templates — no building from scratch
• Hermes Agent handles customer service via Telegram automatically; the GitHub repo hit nearly 100,000 stars faster than any AI agent in history
• Claude Code now works in Excel, PowerPoint, Word and Outlook — I don’t make client reports anymore
• Smartlead sends ~500 cold emails daily; 12-100 replies, ~28% convert to sales
• The critical fix: changing “AI system” to “assistant” in my messaging — people fear AI, not assistants
• It didn’t work on day 1. Three days of silence. The word change on day 4 brought the first sale

Step 1: Paperclip as Your AI CEO

I started with Paperclip — and I want you to notice the words on their site: “Zero Human Company.” That’s not hype. It’s the actual structure. Paperclip becomes your CEO, and underneath it runs a team of agents: a graphic designer, a developer, a marketer, customer service — all working 24 hours while you sleep.

The beauty is the 167 ready-made company templates. You don’t build from zero. You pick a template like a freelancer profile, or a restaurant setup where everything inside is already prepared. The agent I use is called Hermes Agent. I was the first in Turkey to cover it last month in a video about the “crazy AI computer” I got my hands on.

Paperclip manages the company through Claude. Your CMO can be OpenAI, your CTO can be Cursor, your CEO can be Claude, engineering can run on Codex — the system shows you that all of it can be automated. What used to cost me nearly $140/hour in human salaries now stays in my pocket.

Step 2: Hermes Agent for 24/7 Customer Conversations

Hermes is what sits under Paperclip as your customer representative. Here’s why this matters: the GitHub repository has nearly 100,000 stars — no AI agent in history has grown this fast. Why? Because it’s open source, free, and it actually works.

Here’s the loop: Hermes connects to Telegram. A customer writes. Hermes responds automatically. Paperclip generates the report. I check Telegram while walking outside and approve or adjust — all in seconds. I now manage my business on my walks. It’s the best thing I’ve done for myself.

The learning capability is what separates it from basic chatbots. Hermes keeps learning from interactions, doesn’t repeat the same mistakes, doesn’t ask the same questions twice. It runs completely open-source on your own computer, handling everything from video editing to whatever you can imagine. As of now, nothing has surpassed this — though something will, because something new comes every day.

Step 3: Claude Code as Your Production Engine

This is where the actual client work happens. Claude Code builds websites, automations, content — whatever you’re selling. A quick note: I covered how to run Claude Code for free with LLM models in a previous video for anyone who wants that route.

Remember I said limits doubled? Anthropic’s official announcement means I’m working twice as fast now. I deliver to clients in half the time. And on May 7 — two days ago — Claude expanded into Excel, PowerPoint, Word, and Outlook. I don’t prepare Excel reports for clients anymore. Claude makes the PowerPoint presentation and sends it.

I originally built my one-person AI company on Claude alone. That video became the most-asked-about topic on this channel, which is why I’m going deeper now. And realistically, I’m one step beyond even that video today — because the Paperclip + Hermes + Claude Code trio is the formula for transitioning from company to agency.

Step 4: Smartlead for Client Acquisition

AI means nothing without customers. This is critical. Smartlead is a cloud email sending tool that costs roughly $39/month to start. It sends nearly 500 emails daily for me. I get 12 to 100 replies, and approximately 28% of those convert to sales.

Hermes integrates with Smartlead — Hermes is the brain, Smartlead is the hand. For payments, I use Stripe. I send the link, the client clicks and pays, I deliver the service, Stripe notifies me. Some days I’ve done over £16,000 in sales. I covered the full Stripe setup in my previous video where I did £3,000 in 7 days.

Step 5: Bluehost VPS to Keep Everything Running 24/7

Your laptop won’t work — close it and the AI stops. You need a server running 24/7. I use Bluehost VPS. The starter plan at roughly $5-10/month is enough. That’s $5/month, server stays on while you sleep.

My total monthly cost breakdown:

• Smartlead: ~$39
• Bluehost VPS: ~$5
• Claude: ~$20 average
• Total: roughly $40/month

I used to spend over £400 on Claude, ChatGPT, and Cursor. Now I save nearly $300/month.

The One Thing You Must Hear Before Starting

I promised you this in step five. If you’re starting this, don’t start without hearing it: this system doesn’t work on day one. I worked three days. No response. You won’t quit. Then I changed something: word choice. I said “AI system” — I changed it to “assistant.” People are scared of AI. They’re not scared of assistants. What you’re seeing now is a system that automates everything you learned in hundreds of Shopify courses. The next day, the first sale came. Then orders kept coming continuously.

This system depends on word choice. Not the tool. Not the method. The word.

FAQ

How much does it cost to run this AI agency per month?

Approximately $40 total: Smartlead (~$39), Bluehost VPS starter plan (~$5), and Claude usage (~$20 average). This replaced over £400/month I previously spent on multiple AI tools.

What is Hermes Agent and why is it significant?

Hermes is an open-source AI agent that handles customer service via Telegram automatically. Its GitHub repository gained nearly 100,000 stars faster than any AI agent in history because it’s free, open-source, and functional. I was the first in Turkey to cover it.

How does the Paperclip + Hermes + Claude Code combination work?

Paperclip serves as CEO managing the business structure with 167 templates. Hermes operates underneath as customer-facing support. Claude Code executes the actual deliverables — websites, automations, reports, presentations. Together they form a 24/7 agency pipeline.

What was the specific change that made sales start coming in?

After three days of no response, I changed my messaging from “AI system” to “assistant.” The first sale came the next day. The tool and method were identical — only the wording changed. People fear AI but accept assistants.

Conclusion

I built this system in five steps, spent roughly $40/month, and replaced over £400 in previous tool costs. The sales hit while I was sleeping because the infrastructure runs 24/7 on a $5 server. But the technical setup is only half the story — the word choice between “AI system” and “assistant” was what actually converted. If you’re in my community, I’ve shared the exact prompts, cold email templates, SMS marketing systems, and live walkthroughs in Turkish. For anyone feeling lost in this rapid pace of change, that’s completely normal — I’ve spent nearly 15-20 years building this experience, and I’m offering free 15-minute calls to map the fastest path forward for your situation.

Watch the full video (in Turkish — English subtitles available):

Tools & Community

• TurkoLister — the AI listing tool I use to turn Amazon products into optimized eBay UK listings in about 60 seconds (from £4.99/month, £1 one-week trial).
• AI & E-commerce Community — my Turkish-speaking community ($19/month) with weekly live sessions.
• Subscribe on YouTube — new experiments every week.

The post I Built a $16K/Day AI Agency in 5 Steps — What Actually Worked appeared first on Digital Market Mentoring.

Here’s what struck me most: I’ve spent three years building systems that earned me 1.5 million points (roughly £1.5 million in revenue). This person burned nearly that much in 30 days. That gap between efficient and catastrophic AI agent deployment is exactly what I want to close in this article.

Key Takeaways

• Token repetition is the silent killer: AI agents re-reading the same files caused the $1.3M disaster
• Code Graph reduces token costs by ~95%: Map your codebase once, query the map instead of re-reading files
• Match the model to the task: Claude Code for complex work ($20-100/mo), Codex for speed ($10-50/mo), Hermes for local/free
• My three-layer system costs $30-50/month: Single tasks → Code Graph memory → full automation with Telegram control
• Three non-negotiable security rules: spending limits, mandatory Code Graph, and model-task matching

Why AI Agents Burn Money: 3 Root Causes

When I dissected that $1.3 million case, I found three predictable failure patterns. I’ve seen solo developers and small teams repeat them constantly.

1. Repetitive File Reading

Every time an agent receives a task, it insists on reading the necessary code fresh. Five minutes later, another task hits the same file. Then another. The same 50 files, hundreds of thousands of tokens each cycle. No memory, no caching, no learning.

2. Maximum Context Window Abuse

Claude Sonnet 4.5 offers a 200,000-token context window. The agents in that case maxed it out every single request. At roughly $1 per 200K tokens, one agent doing this repeatedly becomes expensive fast. Now multiply by 100 agents running in parallel.

3. Parallel Agent Overload

The developer ran 100 agents simultaneously. Each loaded its own full context. Simple math: 100 agents × 200,000 tokens × constant requests = 1.3 million dollars in one month. The system was architecturally sound but economically naive.

“This system didn’t fail—it was too successful at spending money.”

Code Graph: The 95% Cost Reduction Nobody Talks About

Code Graph sits at #1 on GitHub trends for good reason. It solves the repetition problem at its root.

Here’s how it works: instead of agents reading your entire codebase for every task, Code Graph scans your files once and generates a summary map. When an agent needs to know “what does X function do?” it queries the map—not the original 50 files.

The numbers are stark. Without Code Graph: 50 files read, 100,000+ tokens burned, repeated endlessly. With Code Graph: initial mapping pass, then ~5,000 tokens per subsequent query. That’s where my claimed 95% savings come from.

Code Graph is open-source and integrates with Claude Code, Codex, and Hermes. I run it locally for sensitive projects. For my public-facing automation work, I use it as the memory layer that prevents my agents from “forgetting” what they just read.

Choosing Your Agent: Claude Code vs. Codex vs. Hermes

I use three main agent brands, each for specific situations. GitHub’s February announcement that Claude and Codex now integrate directly into repositories made this even more practical.

Claude Code: Quality at a Price

Claude Code is Anthropic’s agent with strong reasoning capabilities and low error rates. I deploy it for complex coding tasks and customer-critical work where mistakes cost more than tokens. Monthly cost: $20-100. Think of it as the senior developer who won’t stop until the goal is complete.

Codex: My Workhorse for Speed

I use Codex most heavily. It’s fast, cheap, and operates smoothly for repetitive tasks and automations. Monthly cost: $10-50. I’ve integrated it with my phone so I can issue commands without sitting at my desk. It connects across my 4-5 computers and executes with approval workflows.

For social media posting workflows specifically, Codex has been incredibly stable and fast. I describe what I need, and it handles notes, scheduling, and execution in seconds.

Hermes Agent: Local and Free

Hermes runs open-source on your own machine. If you’re handling sensitive data, want zero internet exposure, or need zero marginal cost, this is your option. I use Hermes for eBay automation and any workflow involving API tokens I don’t want externalized.

Combined monthly total across all three: $50-200. Annually, that’s roughly £1,000 in service fees versus the $1.3 million monthly burn of the headline case.

Gemini Spark and the 2026 Personal Agent Race

Google’s Gemini Spark represents something bigger than one product. Every major company—Google, OpenAI, Anthropic—has entered the personal AI agent race. I believe this will be 2026’s defining technology wave.

Spark specifically reads your emails (naturally, Google owns Gmail), monitors your calendar, auto-responds, and suggests meeting plans. It follows the same philosophy as OpenHuman, which I also track closely.

OpenHuman operates as a “personal AI superintelligence” that learns from your GitHub, Stripe, calendar, and email patterns without waiting for prompts. It proactively develops itself. I watched it select my daily priorities from my inbox, request my self-hosting preference, then begin autonomous setup.

The race isn’t just about features anymore. It’s about which system can deliver automation faster and more reliably to regular users. Speed of deployment will determine winners.

My 5-Step Security System: $30-50/Month

This is the practical framework I promised. I’ve refined it through three years of building systems that now generate over £20,000 monthly, with single days hitting £16,000 in sales.

Layer 1: Single Tasks (Week 1)

Give agents one task, one prompt, one output. “Write this specific customer response.” Daily cost: approximately $0.30. No context carryover, no memory, just execution.

Layer 2: Code Graph Memory (Week 2)

Add Code Graph so agents remember previous tasks. They stop re-reading the same files. Daily cost rises slightly to ~$0.50, but capability multiplies.

Layer 3: Full Automation (Week 3)

Connect n8n to a Telegram bot. Agents run autonomously. You control everything from your phone. Daily cost: ~$1.00. Monthly: $30. Annual: $365.

The 3 Mandatory Security Rules

1. Set hard spending limits. Configure your API dashboard with a $50-100 ceiling. When hit, everything stops until you manually review and reset. Non-negotiable.

2. Code Graph is mandatory. I mean this literally: do not deploy AI agents without Code Graph. The 95% token savings isn’t optimization—it’s survival.

3. Match model to task complexity. Simple tasks with Claude Sonnet wastes money. Use Haiku—10× cheaper. Complex tasks with Haiku wastes time and produces errors. Task-appropriate model selection is a discipline, not a preference.

I use OpenRouter’s leaderboard to validate my choices. Hermes currently leads both “top agent coding” and “top proactive” categories, with Claude Code at #4. But “most popular” doesn’t mean “right for you.” DeepSeek V4 Flash might be cheapest for your specific input/output ratio. Analyze the per-million-token pricing yourself.

“Sometimes everyone’s right answer is your wrong answer. Your use case determines your truth.”

FAQ

What exactly caused the $1.3 million AI agent bill?

Three factors combined: agents repeatedly re-reading identical files, maxing out 200,000-token context windows on every request, and running 100 agents in parallel with no shared memory. The system functioned correctly but had no cost controls.

What is Code Graph and why does it save 95% on tokens?

Code Graph is an open-source tool that creates a searchable map of your codebase or documents. Instead of agents reading full files for every task, they query the pre-built map. First scan: full read. Subsequent queries: ~5,000 tokens instead of 100,000+.

How much does your AI agent system actually cost per month?

My three-layer system costs approximately $30-50 monthly: $0.30/day for single tasks, $0.50/day with Code Graph memory, $1/day for full automation via Telegram. My total across Claude Code, Codex, and Hermes ranges $50-200/month depending on workload.

Which AI agent should I choose for my specific needs?

Use Claude Code for complex, error-sensitive work ($20-100/month). Use Codex for fast, repetitive automation ($10-50/month). Use Hermes for local processing of sensitive data (near-zero cost). Match the model to task complexity—never default to the most powerful option.

Conclusion

The gap between $1.3 million monthly burn and my $30-50 system isn’t about having less ambition. It’s about architectural discipline. Code Graph eliminates repetition. Model matching prevents overpayment. Spending limits prevent surprises.

I’ve built systems generating £20,000+ monthly and £1.5 million total over three years. The same tools that could bankrupt a careless operator can empower a methodical one. The technology has arrived. Your economic survival depends on how you deploy it.

My next deep-dive will cover Code Graph installation step-by-step. If you’re building AI agent systems—or trying to avoid becoming the next headline—start with these three security rules before you add a single agent.

Watch the full video (in Turkish — English subtitles available):

Tools & Community

• TurkoLister — the AI listing tool I use to turn Amazon products into optimized eBay UK listings in about 60 seconds (from £4.99/month, £1 one-week trial).
• AI & E-commerce Community — my Turkish-speaking community ($19/month) with weekly live sessions.
• Subscribe on YouTube — new experiments every week.

The post How I Run 100 AI Agents for $30/Month (Not $1.3M) appeared first on Digital Market Mentoring.