> ## Documentation Index > Fetch the complete documentation index at: https://docs.callivox.com/llms.txt > Use this file to discover all available pages before exploring further. # Tools and scopes > The six access scopes a connected AI client can be granted, who can grant each one, and every action the assistant can take. A connected AI client can only do what you approved on the consent screen. Access comes in six **scopes**, and each scope unlocks a specific set of actions — nothing more. ## The six scopes | Scope | What it unlocks | Who can grant it | | ----------------------------- | ----------------------------------------------------------------------------------------------------- | ---------------- | | **Read conversations** | View message threads and contacts. | Admin or Member | | **Read channels** | See connected WhatsApp, Telegram and other channels and their status. Credentials are never returned. | Admin only | | **Read AI agents** | View AI agent configuration. | Admin only | | **Create & manage AI agents** | Create, update and delete AI agents, and read their full configuration. | Admin only | | **Read billing** | See plan, usage and balance. | Admin only | | **Read automations** | View automation flows and their run history. | Admin only | **Read conversations** is the only scope a Member can grant. Everything else needs an Admin. On a Member's consent screen the Admin-only rows appear locked, with an **admin only** chip, and are not granted even if the client asked for them. Scopes are all read-only except **Create & manage AI agents**. Without that scope, a connected assistant cannot change anything in your workspace. ## What the assistant can actually do Each scope maps to a set of named actions — "tools" in AI-client language. You will occasionally see these names in your client's activity log, so here is what each one is. ### Workspace and identity Available to every connection, no scope required. | Tool | What it does | | -------- | --------------------------------------------------------------------------------------------------------------- | | `ping` | A health check that confirms the connection is alive. | | `whoami` | Reports which workspace the connection points at and exactly which scopes it holds. Assistants call this first. | ### Channels — requires **Read channels** | Tool | What it does | | --------------- | ------------------------------------------------------------------------------------ | | `list_channels` | Lists your connected messaging channels and their status. Never returns credentials. | ### AI agents — requires **Read AI agents** | Tool | What it does | | ------------- | ------------------------------- | | `list_agents` | Lists your AI agents. | | `get_agent` | Returns a summary of one agent. | ### AI agents — requires **Create & manage AI agents** | Tool | What it does | | ------------------ | ------------------------------------------------------------------------------------- | | `get_agent_detail` | Returns an agent's full configuration. An assistant reads this before making changes. | | `create_agent` | Creates a new AI agent. | | `update_agent` | Updates an existing agent. | | `delete_agent` | Deletes an agent. Destructive and irreversible. | These four are the only write actions in the whole set, and they act on live configuration immediately — an agent created, changed or deleted by an assistant is created, changed or deleted for real. Ask an assistant to show you an agent's current configuration before you ask it to change anything. An assistant creating an agent can set the same things you would set in the app, including the **Effort** level — the low, medium or high tier that decides how much reasoning the agent applies and how many credits a conversation costs. It cannot pick a provider or a model; the platform handles that. See [Creating an agent](/ai-agents/creating-an-agent). ### Conversations — requires **Read conversations** | Tool | What it does | | --------------------------- | ------------------------------------------------------ | | `list_conversations` | Lists your conversations, most recently updated first. | | `get_conversation` | Returns one conversation. | | `get_conversation_messages` | Returns a conversation's messages, oldest first. | ### Billing — requires **Read billing** | Tool | What it does | | --------------------- | ----------------------------------------------------------------- | | `get_billing_balance` | Returns your credit balance and monthly allowance. | | `get_usage` | Returns your automation-run usage for the current billing period. | | `list_transactions` | Lists recent credit ledger entries, newest first. | ### Automations — requires **Read automations** | Tool | What it does | | --------------------- | -------------------------------------------------------------------- | | `list_automations` | Lists your automations and whether each is published and enabled. | | `get_automation_runs` | Lists recent automation runs, newest first, optionally for one flow. | ## Things worth knowing * **Long lists arrive in pages.** An assistant asking for hundreds of conversations receives them in batches and fetches the next page as needed. You may see it work through a large result set in steps. * **A missing scope is a clean refusal, not a failure.** Calling something outside the granted scopes returns a "forbidden" answer, and a well-behaved assistant will tell you it lacks the access rather than retrying. * **Everything is limited to your own workspace.** There is no way for a connected client to reach another customer's data. * **Scopes cannot be edited after the fact.** To change what a client may do, revoke it and connect again — see [Connecting an AI client](/mcp/connecting-an-ai-client). ## Next steps Setting up a connection and revoking it later. What MCP is for and where the page lives.