Beyond the core Knowledge, Behavior, and Design tabs, there are a few more tabs worth knowing about.
Settings tab
Go to your agent → Settings tab.
Agent name — you can rename the agent here.
Language — set the primary language your agent should respond in. The default is English. The agent will still understand questions in other languages but will respond in the configured language.
Reply mode — controls how the agent handles new messages:
- Auto-reply — the agent sends responses immediately without any human review
- Draft — the agent writes a response but holds it for a human to review and approve before sending. Useful if you want oversight, especially during the early days of deployment.
Channels tab
This is where you connect the agent to channels — Gmail, Telegram, website widget, etc. Each channel you add here is a separate conversation entry point. See Channels overview for details.
Tools tab
Tools are integrations that let your agent do things beyond just answering questions — like looking up order status, checking a calendar, or creating a support ticket. If you have custom tools connected, they'll appear here.
Helpdesk tab
If you're using Vosaire alongside a helpdesk like FreeScout, you can configure escalation behavior here. When a conversation needs a human, the agent can route it to your helpdesk automatically.
Contacts (Leads) tab
Vosaire can capture lead information during conversations — name, email, custom fields. The Contacts tab shows all the leads this agent has collected. You can export them or view conversation history for each contact.
Insights tab
Agent-specific analytics: conversation volume over time, resolution rate, most common topics, and user satisfaction scores for this specific agent. Useful for comparing agent performance if you have multiple.
Train tab
The Train tab shows conversations where the agent gave incorrect or incomplete answers and a human marked them for correction. You can review and fix the responses here — each correction gets added to the agent's knowledge base as a high-priority FAQ entry.
This is one of the best ways to improve agent quality over time. Real user questions, real failures, real fixes.