Skip to content
↑↓Navigate↵SelectescClose

Frequently Asked Questions

Common questions about using Nexus

It depends on your role:

If you’re a User (making AI requests, viewing your own usage on an already existing platform):

  • Start with Getting Started to create your first API key
  • Then read Using Nexus to understand classification, sessions, and model selection
  • Check the FAQ for common questions

If you’re a Developer (integrating Nexus into your applications):

If you’re a Manager (managing team members, setting budgets, assigning roles):

If you’re an Auditor or Compliance Officer (reviewing audit logs, generating compliance reports):

If you’re in Finance or Billing (monitoring costs, reconciling spending):

If you’re in Security or InfoSec (reviewing security controls, configuring policies):

If you’re a Platform Administrator (deploying Nexus, configuring SSO, managing infrastructure):

If you’re an Executive or Decision Maker (understanding what Nexus does and why):

  • Start with Welcome for a high-level overview and capabilities
  • Read Getting Started to see how Nexus solves governance challenges

Contact your administrator. They control who has access to Nexus. They can add you to the right organization and project.

No. You can use the Nexus console to view usage, manage keys, and monitor activity without writing code. However, to actually send AI requests from your applications, a developer will need to integrate Nexus.

Yes. Nexus encrypts API keys and stores them securely. Only people in your organization with proper access can see them. Always keep your keys private and never share them publicly.

You cannot recover a lost key, but you can delete the old one and create a new one. Update your applications to use the new key.

Nexus itself does not charge you—your costs come from the AI providers (OpenAI, Anthropic, etc.). Nexus just tracks and displays those costs. Talk to your administrator about your organization’s billing model (BYOK or PTB).

Your administrator can set spending limits. If you hit a limit, Nexus will stop accepting requests until the limit is raised or the time period resets. Check with your admin to adjust your budget.

Yes. Nexus supports many models from different providers. You can switch between GPT-4, Claude, and others. Each model has different pricing and capabilities.

What is the difference between Workspace, Organization, and Project?

Section titled “What is the difference between Workspace, Organization, and Project?”

Nexus organizes access in three levels. A Workspace is your top-level account (usually one per company). An Organization is a team or department within your workspace (like Engineering, Product, or Marketing). A Project is a specific application or use case within an organization (like a customer support chatbot or documentation assistant). Each project gets its own API keys and usage tracking. Learn more about workspace structure.

Why does Nexus have different levels (Workspace/Org/Project)?

Section titled “Why does Nexus have different levels (Workspace/Org/Project)?”

This hierarchy provides security isolation and cost tracking at different scales. You can answer questions like “How much is the Marketing team spending on AI?” or “Which application is using the most tokens?” You can also set different security rules for different use cases—for example, your customer support chatbot can handle confidential data while your marketing content generator only handles public information. Each project is isolated so a compromised API key in one project cannot access data from another project. Learn more about security benefits.

Data classification is how you label the sensitivity level of each request. Every request can carry a classification tag like public, internal, confidential, or restricted. Your developer sets this label using the nexus-classification header when making requests. Nexus uses these labels to enforce security policies, control which projects can handle sensitive data, and create audit records showing how data was processed. Learn more about classification.

Does classification prevent sensitive data from being sent to OpenAI/Anthropic?

Section titled “Does classification prevent sensitive data from being sent to OpenAI/Anthropic?”

No. Classification does not change what gets sent to the AI provider—the full request (including any sensitive data) is forwarded to OpenAI or Anthropic. However, classification does control important security policies: it blocks requests that exceed your project’s classification ceiling (for example, if your project can only handle internal data, requests labeled confidential are rejected before reaching the provider), it determines how data is logged in Nexus’s audit trail (for example, emails might be redacted to ***@*** in logs), and it controls which AI providers are approved to handle that classification level. Learn more about what classification controls.

Can I change classification during a conversation?

Section titled “Can I change classification during a conversation?”

Yes. Your developer can set a different classification for each request, even within the same conversation. For example, a customer support chatbot might send one request with nexus-classification: internal when asking about business hours, then send the next request with nexus-classification: confidential when the customer provides payment information. The developer’s code detects the change in data sensitivity and adjusts the classification header accordingly. For technical details on how to implement this, see the Nexus Developer documentation.

Can AI agents and LLMs access this documentation?

Section titled “Can AI agents and LLMs access this documentation?”

Yes. This documentation is optimized for AI agent consumption following the llms.txt v2 specification. AI agents can:

  1. Discover content via /llms.txt which provides a structured overview of all documentation
  2. Fetch clean markdown at any page URL by adding .md to the path (e.g., https://docs.vnexus.ai/home/features-overview.md)
  3. Find everything via /robots.txt which references both the sitemap and llms.txt

This makes it easy for coding agents, AI assistants, and automated tools to read and understand Nexus documentation without parsing HTML. All markdown files are automatically generated during the build process with no manual maintenance required.

Give your AI agent the llms.txt URL as a starting point:

https://docs.vnexus.ai/llms.txt

This file provides a complete overview with links to all major documentation sections. Your agent can then follow links to specific markdown pages as needed. All page URLs work in both HTML (for humans) and markdown (for agents) formats.

  • Check this documentation for guides and reference materials
  • Ask your administrator for access and billing questions
  • Contact your IT team for technical integration help
  • Open an issue on the Nexus GitLab repository for bugs or feature requests