> ## Documentation Index
> Fetch the complete documentation index at: https://docs.sontairo.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Core Concepts

> Understand the building blocks of Sontairo

Sontairo is organized in two layers: the **products** that run parts of your business (Hosting, Helm CRM, Finance, Voice), and the **machinery** they share (workspaces, projects, agents, connectors).

## Workspaces

A workspace is the top-level container for your organization. Every resource — products, agents, connectors, projects, runs — belongs to a workspace.

* Each user can belong to multiple workspaces
* Workspace members share access to all resources within it
* Billing, subscriptions, and product add-ons are managed per workspace: one active add-on covers the whole account

## Projects

Projects are logical boundaries within a workspace for organizing work. Data is **fully isolated** between projects:

* Agents, connectors, automations, chat conversations, and runs are scoped to a project
* Switching projects updates all views to show only that project's data
* A resource with no project belongs to the workspace globally

## Products

| Product                          | What it runs                                                                                        | Pricing                |
| -------------------------------- | --------------------------------------------------------------------------------------------------- | ---------------------- |
| **[Hosting](/hosting/overview)** | Your website — built by AI or deployed from a folder, the CLI, or GitHub, with AI health monitoring | Monthly add-on         |
| **[Helm CRM](/helm/overview)**   | Your customer data — bases, tables, views, records, and automations                                 | Monthly add-on         |
| **[Finance](/finance/overview)** | Your payments — payment links and site checkout, with in-app payouts                                | Free (2% platform fee) |
| **[Voice](/voice/overview)**     | Your calls and texts — voice mode in chat and an AI-answered phone number                           | Monthly add-on         |

Ultra and Enterprise plans include all paid products. See [Plans & Pricing](/billing/plans).

## Agents

Agents are AI-powered assistants configured with:

* **System Prompt** — Instructions that define the agent's behavior
* **Provider Key** — The AI model and API key used for inference
* **Connectors** — External tools the agent can access during task execution

Agents can be used in chat conversations, called from an automation, triggered through tasks, or run by [Helm automations](/helm/automations).

## Connectors

Connectors integrate Sontairo with external services. They support:

* **OAuth 2.0** — Secure token-based authentication (Google, Slack, GitHub, etc.)
* **API Keys** — Direct key-based access for services without OAuth
* **MCP (Model Context Protocol)** — Custom tool servers for advanced integrations

Each connector belongs to a workspace and optionally a project.

## Tasks & Runs

* **Task** — A one-off instruction sent to an agent, optionally with connectors attached
* **Run** — The execution record of a task, including status, logs, and output

Runs are tracked in the Runs timeline with detailed step-by-step visibility.

## Automations

Build reliable workflows by connecting a manual or scheduled trigger to logic, transformations, AI tasks, delays, connectors, webhooks, and notifications.

* Resolve configuration and access questions before activation
* Publish immutable versions while keeping later edits in a draft
* Pause schedules without deleting them
* Inspect every run and component attempt

## Chat Conversations

Interactive sessions with AI agents featuring:

* Streaming responses
* Persistent conversation history
* Agent memory across conversations
* Tool use during chat (via connected tools)
* Hands-free [voice mode](/voice/overview) on Voice-enabled plans

## Approvals

Human-in-the-loop workflows that pause agent execution until a user approves or rejects the proposed action. Approvals are workspace-scoped.