YecoAI
Platform

DeskNexo

DeskNexo is the AI support agent for hosting providers: it reads WHMCS tickets, diagnoses VMs on Proxmox and VirtFusion, resolves the issue, and writes a human reply to your customer, alone, in under 30 seconds, 24/7. Every action is snapshot-protected and dual-LLM verified.

The same tickets every day, resolved in seconds: one agent that reads WHMCS tickets, fixes the VM on Proxmox or VirtFusion, and replies for you. From €150/month, WHMCS plugin included.

How DeskNexo works, end to end

A customer opens a ticket in your WHMCS at 3 AM. DeskNexo reads it in real time, classifies the request, and queries the hypervisor through the native API of Proxmox or VirtFusion (Virtualizor supported as a control panel) to see the actual VM state: running processes, disk usage, snapshot history, console output. It diagnoses the problem, executes the fix within the action whitelist you define, drafts a clear human reply, and closes the ticket with a summary of what was done. Median resolution time: under 30 seconds, on a single ticket, without waking anyone up.

Two modes: Copilot and Autopilot

Copilot mode prepares diagnosis and resolution, and waits for your approval before anything is sent: ideal while you build trust or for high-risk operations. Autopilot mode works the queue alone, end to end, within the boundaries you set: allowed VM actions, escalation rules, tone of voice. You can switch modes per ticket category, run Autopilot at night and Copilot during business hours, and take over any conversation mid-flight. The agent is never a black box: every decision is logged and explainable.

Dual-LLM verification, not just a prompt

Every ticket follows a strict pipeline: a generator model reads the ticket and the panel context and proposes a resolution; a second, independent verifier model judges the proposal against the customer's original request before anything is sent. If the verifier rejects, the answer never reaches the customer: it escalates to a human with full context. Hallucinated answers stop at the gate, not in your review queue. The same architecture guards destructive operations: reboot, resize, password reset and snapshot restore always require an explicit rule or your approval, and a snapshot is taken before every state-changing action.

Compliance built in, not bolted on

GDPR Guard is active on every conversation: customer messages are masked before they reach any model, with PII detection powered by our own Yeco-PII model running on European infrastructure. PII detection relies on Yeco-PII, described in the 2.0 research. Training data never leaves your tenant, and no conversation is used to train shared models without your explicit opt-in. Every read, proposal, action and reply is written to an immutable audit trail, exportable for compliance reviews. Your customers' data stays in the EU, under your policy.

Setup in 30 minutes

No migration, no rewrite, no new panel to learn. Step one: install the WHMCS plugin, about 10 minutes, no code. Step two: connect the Proxmox or VirtFusion API token and define which VM actions are allowed, about 20 minutes. Step three: let it work, and watch the first tickets resolve from the dashboard. Overage is billed transparently, with alerts at 80% and 100% of your monthly limit, and every number (resolution rate, cost per ticket versus a human agent, SLA, token spend) is exportable.

Plans that match your queue

Essential (€150/month) includes 200 AI tickets with Autopilot and Copilot plus read-only VM diagnostics, overage at €0.10 per ticket. Business (€400/month) raises the queue to 800 tickets and unlocks all VM actions, overage at €0.08. Enterprise (€800/month) includes 5,000 tickets, custom configuration, a self-hosted LLM option for full data control, overage at €0.04. Pricing is per ticket, never per agent seat: your whole team benefits at the same cost. The Beta program is limited to three verified hosting companies, with onboarding and direct access to our engineers.

Built for the hosting economy

Hosting support math is unforgiving: 80% of tickets are repetitions, 40 hours a week lost to L1 triage, 3x staffing for 24/7 coverage, 12% churn after slow replies. DeskNexo takes over exactly that work: the repetitive first line, end to end, at a fraction of the cost of a night shift, while your team keeps control of everything that matters. The result is measured, not promised: resolution rate, first-response time and cost per ticket are on your dashboard from day one.

Related products