HOW CAN WE HELP?

Find answers, get support, and learn how to make the most of Orchestrator

FREQUENTLY ASKED QUESTIONS

How do I get started with Orchestrator?+

Sign up for an account, complete the onboarding flow, and create your first agent.

What's the difference between agents and skills?+

Agents define how your website assistant behaves. Skills are reusable instruction folders you attach to an agent so it can follow specialized workflows.

How does the approval system work?+

When an agent needs to use a tool, you can require manual approval. The agent will pause and wait for you to approve or reject the tool execution through the UI.

Can I use my own LLM models?+

Currently we support OpenAI (GPT-4, GPT-3.5) and Anthropic (Claude 3) models. Custom model integration is available for Enterprise plans.

How secure are the sandboxes?+

All tool executions run in isolated Docker containers with configurable resource limits, network restrictions, and security policies. Each execution is completely isolated.

What's included in the audit logs?+

The tenant audit log records governance events: who changed agents, skills, deployments, API keys, team access, billing, and tenant policies. Retention depends on your plan (typically 30–365 days; enterprise plans can export CSV). For live runs, use session history and transcript export—those are separate from the audit log and are not a full record of every API call or tool invocation.

STILL NEED HELP?

Send us a message and we'll get back to you within 24 hours