> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://zuora.ferndocs.com/_mcp/server. # Zuora AI > Use Zuora AI to analyze data, investigate issues, and take action across billing, revenue, and collections using natural language. Use Zuora AI to analyze data, investigate issues, and take action across billing, revenue, and collections using natural language. Zuora AI is embedded intelligence built directly into your quote-to-cash system. It helps finance teams investigate issues, understand financial impact, and take action within the controls and workflows they already use. Zuora AI consists of purpose-built agents with specialized skills designed for finance use cases across billing, revenue, and collections. > **Note** > > Zuora AI tenant settings are managed by a OneID Organization Admin. For information about the available modes and how to change your organization's setting, see [Zuora AI modes](/docs/platform/zuora-ai/ai/access-and-permissions-for-zuora-ai). ## Ways to interact with Zuora AI Zuora AI can be accessed through multiple experiences that are integrated into your workflow. **Chat with Zuora AI** - The chat experience provides a conversational interface for interacting with your data. * Side Panel - The side panel offers a persistent, context-aware assistant alongside your workspace, allowing you to ask questions without leaving your current page. * Full Screen - For more complex tasks, full screen mode provides an expanded workspace where you can perform deeper analysis and multi-step interactions. * Runbooks - Zuora AI also supports runbooks, which are reusable workflows that combine multiple steps into a guided process. * Controls - Controls allow administrators to define tenant-level prompts that influence how AI behaves across the organization. * Prompt chips - After each response, Zuora AI provides contextual suggestions to help guide the next step. * Thinking - The thinking view provides visibility into how the AI generates its responses. **Integrated AI** - Integrated AI is embedded directly into product workflows. Narrative summaries are available across Billing, Revenue, and other areas such as accounts, subscriptions, invoices, and payments. These summaries help you quickly understand the state of your data without manual analysis. **Zuora MCP Servers** - Zuora AI can also be accessed through Zuora MCP Servers, which allow supported external tools such as Claude Desktop, Cursor, Windsurf to connect directly to Zuora. This enables use cases such as exploring APIs, generating integration code, querying tenants, running reports, upgrading SDKs, and managing subscriptions from external environments. These interactions follow the same permission model as in-product AI usage. > **Note** > > Zuora AI uses the details available in your prompt, context, conversation history, permissions, and connected tools to determine how to respond. For best results, provide clear and specific instructions, including the business context, expected output, relevant records or time periods, and applicable constraints. If a response does not match your intended result, refine your prompt with additional details or corrections to help Zuora AI better understand the outcome you want. If behavior remains incorrect even with a clear and detailed prompt, contact Zuora Global Support for further investigation. > Use Zuora AI to analyze data, investigate issues, and take action across billing, revenue, and collections using natural language. ## Docs - [Tenant access to AI](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/access-and-permissions-for-zuora-ai.md): Understand how Zuora AI access is controlled by tenant settings, user permissions, and organizational guardrails. - [Watch Tower and AI activity logs](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/watch-tower-and-ai-activity-logs.md): Watch Tower provides administrative visibility, auditability, and human approval for Zuora AI activity. - [Usage guidelines for Zuora AI](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/usage-guidelines-for-zuora-ai.md): Zuora AI limits and constraints help you plan batch updates, file uploads, and revenue reports within supported request and workflow boundaries. - [Controls for Zuora AI](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/controls-for-zuora-ai.md): Understand how tenant-level Controls define and enforce guardrails for Zuora AI, preventing AI-assisted actions that conflict with your organization’s business policies. - [Approval Control](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/controls-for-zuora-ai/approval-control.md) - [Control properties](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/controls-for-zuora-ai/controls-info.md): Reference details for Control fields and statuses. - [Audit trail for Zuora AI](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/audit-trail-and-conversation-history.md): Zuora records AI-assisted actions and conversations for audit and compliance, providing traceable audit entries and accessible conversation history. - [AI Source and AI SubSource in the Audit Trail](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/audit-trail-and-conversation-history/ai-source-and-ai-subsource-in-the-audit-trail.md): Use the AI Source and AI SubSource fields in Audit Trail logs to determine whether a change was initiated through AI and trace it to the originating interaction. - [Manage files in Zuora AI](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/files-handling-in-zuora-ai.md): Learn how to upload, generate, and download files in Zuora AI. - [Runbooks](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/runbooks.md): Use Runbooks to automate repeatable, multi-step AI workflows for common tasks such as account reviews, payment investigations, and configuration support. - [Create Runbooks](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/runbooks/create-runbooks.md): Create reusable Runbooks in Zuora AI from scratch or by saving existing conversations. - [Zuora AI Chat features and capabilities](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/function-calls.md): Zuora AI shows how it processes your requests, including function calls, thinking steps, and visual summaries, so you can understand the source, structure, and presentation of its responses. - [Zuora AI prompting and capabilities](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/zuora-ai-prompting-and-capabilities.md): Zuora AI responds to natural-language requests about data in your Zuora environment. The quality of the response depends on how clearly you state the outcome, the records involved, and the scope. - [Write an effective Zuora AI prompt](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/zuora-ai-prompting-and-capabilities/write-an-effective-zuora-ai-prompt.md) - [Zuora AI prompt pattern](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/zuora-ai-prompting-and-capabilities/zuora-ai-prompt-pattern.md): Reusable prompt patterns for Zuora AI, with example prompts for account, payment, revenue, and workflow requests. Replace bracketed values with information from your tenant. - [Zuora AI agent areas](https://zuora.ferndocs.com/docs/platform/zuora-ai/ai/zuora-ai-prompting-and-capabilities/zuora-ai-agent-areas.md): Business areas covered by Zuora AI agents and representative uses for each. Availability depends on your Zuora products and enabled capabilities.