Question 1

What is Nexaera?

Accepted Answer

Nexaera is the world's first Vibe-Code Agent Builder. Instead of dragging boxes or writing code, you describe the AI agent you need in plain English — and our AI assistant Freddie builds it for you in real time. You say what you want, Freddie creates the agent, configures its skills, and deploys it. That's it.

Question 2

What does 'Vibe-Code' mean?

Accepted Answer

Vibe-Code means you build by describing, not by configuring. You tell Freddie something like 'Build me a customer support agent that handles refunds and checks order status' — and it creates the agent, writes the system prompt, enables the right skills, and sets up memory. You refine through conversation: 'Make it more formal' or 'Add calendar booking.' The agent evolves through chat, not config screens.

Question 3

Who is Nexaera for?

Accepted Answer

Nexaera is for anyone who wants to deploy AI agents without writing code or wiring together tools manually. That includes digital agencies selling AI services to clients, consultants automating workflows for businesses, SaaS founders adding AI capabilities to their products, and small business owners who want an AI employee handling real tasks — support, scheduling, lead follow-up, research.

Question 4

Do I need technical skills to use Nexaera?

Accepted Answer

No. There is no code, no terminal, no config files. You have a conversation with Freddie, describe what you need, and the agent is built. If you can describe a job role to a new hire, you can build an agent on Nexaera.

Question 5

What can agents actually do?

Accepted Answer

Agents on Nexaera can: answer customer questions via chat or voice, send and read emails, post messages to Slack, manage Google Calendar events, search the web and summarize results, execute multi-step workflows with conditional logic, remember context across conversations, run autonomously on schedules (daily reports, hourly monitoring), answer phone calls, and deploy as chat widgets on any website. All of this is configured through conversation with Freddie — no manual setup.

Question 6

How is Nexaera different from ChatGPT or Claude?

Accepted Answer

ChatGPT and Claude are chat interfaces for general AI conversations. Nexaera creates autonomous agents that do real work — they have tools (email, calendar, Slack, web search), memory, scheduled tasks, and multi-channel deployment. A Nexaera agent doesn't just answer questions — it takes actions, follows up, and operates independently.

Question 7

How is Nexaera different from Zapier or Make.com?

Accepted Answer

Zapier and Make run predefined automations: 'when X happens, do Y.' Nexaera agents reason and adapt — they understand context, make decisions, and handle situations that don't fit a rigid if/then flow. They're AI employees, not automation recipes.

Question 8

Who is Freddie?

Accepted Answer

Freddie is your AI co-builder inside Nexaera. When you open the agent builder, Freddie guides you through creating an agent by asking what you need and building it in real time. Freddie handles the system prompt, skills, memory, voice settings, and deployment — all based on your plain-English instructions.

Question 9

How does the Vibe-Code builder work step by step?

Accepted Answer

Three steps: (1) Describe — tell Freddie what agent you need, like 'A voice agent for my dental clinic that books appointments.' (2) Freddie builds — it creates the agent, writes the system prompt, enables skills (email, calendar, web search, etc.), configures memory, and sets voice if needed. You watch it happen in real time. (3) Deploy and iterate — your agent is live instantly. Want changes? Just tell Freddie: 'Add a Spanish greeting' or 'Make it ask for insurance info first.'

Question 10

Can I refine an agent after it's created?

Accepted Answer

Yes, that's the whole point. Vibe-Code means iterative refinement through conversation. You can tell Freddie to change the tone, add or remove skills, adjust the system prompt, update welcome messages, or reconfigure voice settings — all without leaving the chat. Every change is applied immediately.

Question 11

What skills can Freddie give an agent?

Accepted Answer

Freddie can enable 23+ skills including: send email, read email, web search, create/read calendar events, post to Slack, read/write memory, make outbound calls, and connect to any external API via MCP connectors. Skills are enabled based on what you describe — you don't need to know the skill names.

Question 12

Can I still configure agents manually?

Accepted Answer

Yes. Every agent has a full settings panel where you can manually edit the system prompt, toggle individual skills, adjust voice settings, configure memory, set up scheduled tasks, and manage deployments. Freddie is the fastest way to build, but you always have full manual control.

Question 13

What AI models power the agents?

Accepted Answer

Nexaera supports models from OpenAI (GPT-4o, GPT-4o-mini), Anthropic (Claude 3.5 Sonnet, Claude 3 Haiku), and Google (Gemini 2.5 Pro, Gemini 2.5 Flash). You can set a primary model and a fallback model per agent. Each organization provides their own API keys for whichever providers they prefer.

Question 14

What is agent memory?

Accepted Answer

Memory lets agents remember information across conversations — customer preferences, past interactions, key facts. When a user returns, the agent knows their context without being told again. Memory is scoped per-agent and stored securely. You can view, edit, or clear an agent's memory at any time.

Question 15

Can agents run on a schedule without anyone triggering them?

Accepted Answer

Yes. Heartbeat tasks let you configure agents to wake up on a schedule — every few minutes, daily at a set time, or on a cron expression — execute a task (like 'check for new support tickets and summarize them to Slack'), log the results, and go back to sleep. Smart scheduling adjusts frequency when nothing has changed.

Question 16

Where can I deploy an agent?

Accepted Answer

Agents can be deployed as: (1) a website chat widget embeddable on any site, (2) a hosted chat page with its own URL, (3) a WhatsApp bot, (4) a Telegram bot, (5) a voice agent that answers phone calls, or (6) a workflow endpoint triggered by webhooks. Each channel is configured independently.

Question 17

Can agents handle voice and phone calls?

Accepted Answer

Yes. Nexaera has built-in voice capabilities — agents can have real-time voice conversations in the browser, and they can be configured with phone numbers to answer inbound calls or make outbound calls. You configure the voice (provider, voice ID, speed, stability) per agent.

Question 18

What is the Visual Workflow Builder?

Accepted Answer

The workflow builder is a drag-and-drop canvas for creating multi-step agent pipelines. You connect nodes — triggers, LLM reasoning steps, conditions, API calls, email, Slack, calendar, delays — into complex automations. Workflows can be attached to agents as their execution logic, enabling multi-stage processes without code.

Question 19

Can I white-label Nexaera for my clients?

Accepted Answer

Yes. You can apply your own brand name, logo, colors, and custom domain. Your clients see your brand — not Nexaera. There's no 'Powered by' badge in the client-facing experience.

Question 20

How do client sub-accounts work?

Accepted Answer

Each client gets an isolated workspace with their own agents, conversations, API keys, and usage limits. No client can see another client's data. The isolation is enforced at the database level. You manage all clients from a single agency dashboard.

Question 21

Can clients log in and manage their own agents?

Accepted Answer

Yes. You can invite client team members with role-based permissions. Clients see only their own workspace. You can also manage everything on their behalf — the platform supports both models.

Question 22

Can I set different permissions per client?

Accepted Answer

Yes. Per sub-account you can control which features are accessible, the maximum number of agents, monthly task run limits, and whether clients can manage their own API keys. This lets you offer different service tiers.

Question 23

What integrations are available?

Accepted Answer

Built-in integrations include: Email (SMTP/IMAP), Gmail (with push notifications), Google Calendar, Slack, Webhooks (inbound and outbound), and Web Search. These are full bidirectional integrations that agents use as tools during execution — not just data connectors.

Question 24

What is MCP and how do custom connectors work?

Accepted Answer

MCP (Model Context Protocol) is an open standard for connecting AI agents to external tools. You can add MCP connectors to give agents access to any third-party API — a CRM, a database, a custom service. Once connected, agents can call those tools just like built-in integrations. MCP connectors support API key, OAuth, and custom header authentication.

Question 25

Are integration credentials shared between clients?

Accepted Answer

No. Every integration is configured per sub-account. Each client has their own OAuth connections, their own credentials, and their own usage logs. No cross-tenant access.

Question 26

How are credentials stored?

Accepted Answer

All API keys, OAuth tokens, and access credentials are encrypted at rest before storage. They are never stored in plaintext and are only decrypted at execution time inside backend functions. Token expiry is monitored and you're notified before tokens expire.

Question 27

How is client data protected?

Accepted Answer

Client data is protected through: database-level tenant isolation with row-level security, encrypted credential storage, audit logs for every action, role-based access controls, and complete data separation between tenants. No client data is used for model training.

Question 28

What are Approval Gates?

Accepted Answer

Approval Gates are pause points where an agent stops and asks a human to approve an action before proceeding — like sending an email, processing a refund, or calling an external API. This prevents AI mistakes from having real-world consequences. You configure which actions require approval per agent.

Question 29

What are Audit Logs?

Accepted Answer

Audit logs record every significant action: agent executions, tool calls, user logins, configuration changes, and approval decisions. Each entry includes timestamp, actor identity, action type, and before/after data. Logs are exportable and retained based on your plan.

Question 30

What happens if an agent makes a mistake?

Accepted Answer

Approval Gates catch high-risk actions before they execute. Every action is logged, giving you a full audit trail. Agents operate with least privilege — they only access the tools and integrations you explicitly configure. Combined, this significantly reduces both the probability and impact of errors.

Question 31

How does pricing work?

Accepted Answer

Nexaera charges a flat monthly platform fee based on your plan. There are no per-agent fees or per-message fees. LLM API usage is billed by the AI provider (OpenAI, Anthropic, Google) based on the API keys configured in your account — Nexaera does not mark up AI usage. Visit our pricing page for current plan details.

Question 32

Why do users bring their own API keys?

Accepted Answer

Bring-your-own-key means: (1) you have full transparency on AI costs through your provider's dashboard, (2) there's no markup on token usage, (3) you choose which models to use, and (4) Nexaera doesn't become a middleman for AI billing. It keeps costs predictable and transparent.

Question 33

Is there a free trial?

Accepted Answer

Yes. All plans include a 14-day free trial with full access to every feature on your selected plan. Build agents, configure integrations, and test the platform before committing.

Question 34

What counts as a 'task run'?

Accepted Answer

A task run is one execution of an agent's task — triggered manually, by a schedule, or by a webhook. A single task run can involve multiple tool calls (search the web → summarize → post to Slack), but it all counts as one run.

Question 35

Can I upgrade or downgrade?

Accepted Answer

Yes. You can change your plan anytime from the Billing section. Upgrades take effect immediately. Downgrades apply at the end of your current billing period.

Question 36

Does Nexaera have an API?

Accepted Answer

Yes. REST API endpoints let you trigger agent tasks, post to conversations, read execution logs, manage knowledge sources, and handle webhooks. Authentication uses API keys generated per account.

Question 37

What is the Knowledge Base?

Accepted Answer

The Knowledge Base lets you upload documents (PDF, DOCX, TXT) or web URLs that agents can reference during tasks. Content is chunked, embedded as vectors, and retrieved semantically when relevant — so agents can answer questions about your products, policies, or data without the information being in the system prompt.

Question 38

Does Nexaera support multiple languages?

Accepted Answer

Yes. Agents can be configured to respond in specific languages or auto-detect the user's language. This is per-agent and supports all languages supported by the underlying LLM models (50+). The platform UI is in English.

Question 39

Can I import agents from another platform?

Accepted Answer

Yes. The Import Agent feature lets you bring in agent configurations including system prompts and workflow structures. For complex migrations, you can also use Freddie — just describe what your existing agent does and Freddie will recreate it.

Question 40

What support is available?

Accepted Answer

Support options vary by plan and include email support, priority support queues, and onboarding calls. All users have access to documentation and in-app contextual guidance. Check our pricing page for plan-specific support details.

Question 41

Where can I find documentation?

Accepted Answer

Documentation including quickstart guides, API reference, and integration setup instructions is available in-app. Each section of the dashboard also has inline help tooltips and contextual guidance.

Question 42

Can I request new features?

Accepted Answer

Yes. We maintain a feature request board where you can submit and upvote ideas. We prioritize based on demand, impact, and feasibility.

Nexaera FAQ — pricing, MCP, white-label, and agency plans

What is Nexaera?

What is a sub-account?

What counts as a task run?

What is white-label MCP?

Can I use my own API keys?

Is there a free trial?

Does Nexaera sign a HIPAA BAA?

How is tenant data isolated?

What is the SLA?

Who are the real competitors?