Skip to content
// FAQ

Straight answers, including the unflattering ones.

Every question we get asked more than once, grouped by topic. Where an answer needs real depth it has its own page, and the rest are answered here.

// If your question is not here, ask it and we will answer directly.

// Trust

Safety and data

Where your code goes, what leaves your machine, and what we will not claim.

Is it safe to use third-party tools with Claude Code?

Full answer

It depends entirely on what the tool does with your code. The question worth asking is whether the tool executes your code or only coordinates around it.

Where does my code and data run?

Full answer

Your code runs on your machines. Virex holds coordination state, session records, and the memory your agents write.

Is Virex SOC 2 certified?

No. Controls are designed to a SOC 2-oriented control set and no audit has been completed. The security page states this plainly.

How is one organization isolated from another?

At the database layer, with row-level security, so a query scoped to one organization cannot return another organization data.

// Virex Remote

Running agents remotely

Oversight and control of coding sessions on machines you own.

How do I run Claude Code on my own machine remotely?

Full answer

Keep the session local and add a control plane over it, rather than moving the agent into someone else cloud.

What can I do to a running session?

Follow live activity, send an instruction, approve or deny permission and plan decisions, and interrupt the current activity.

Which coding agent does Virex Remote support?

Claude Code is the primary integration. Remote is built around how its sessions run and pause.

Does my machine need to stay awake?

Yes. The session runs on your hardware, so the machine has to be running and connected.

// Virex Memory

Agent memory

Context that survives the session, and who can reach it.

Does agent memory work across Cursor and Claude Code?

Full answer

Yes. Memory sits behind MCP, so every compatible client reads and writes the same organization corpus.

What gets remembered?

Whatever your agents write as they work. It is a store agents use rather than a transcript recorder.

Does the corpus degrade as it grows?

Relevance scoring is refreshed on a schedule and stale memories expire automatically.

Can teammates read what my agents wrote?

Within your organization, yes. That sharing is the point, and isolation between organizations is enforced in storage.

// Commercial

Accounts, seats, and billing

How the account model and the subscription actually work.

How is Virex priced?

Per seat, per month, per product, at the same flat rate for both products. Current rates are on the pricing page.

Is there a free tier or a trial?

No. Access begins when your organization subscribes and an admin assigns you a seat.

Do we need a subscription per product?

No. One subscription per organization, with one line item per product.

Can a user belong to more than one organization?

No. Each user belongs to exactly one organization.

What happens when someone leaves?

An admin removes their seat, which ends their access. What their agents wrote stays with the organization.

// More detail

Where the longer answers live.

Security

Isolation, identity, audit logging, retention, and the threat model, at mechanism level.

Glossary

Definitions for MCP, agent memory, permission requests, and the rest of the vocabulary.

Integrations

Which coding tools work with which product, and why the two lists differ.

Security, glossary, integrations, and pricing.

Still have a question?

Ask it directly. We answer honestly, including when the answer is that we are not the right fit.