> For the complete documentation index, see [llms.txt](https://docs.limio.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.limio.com/product/mcp-server/mcp-server.md).
# Overview
The Limio MCP Server connects AI assistants like Claude to your Limio environment. Instead of navigating the admin dashboard, you can ask questions and give instructions in plain language.
**Examples of what you can do:**
* "Show me all our published student offers"
* "Create a monthly digital subscription at $9.99"
* "Update the price on our annual plan to $129.99"
* "Where is the Black Friday offer shown?"
* "How many orders came through our student offers this month?"
The server uses the [Model Context Protocol](https://modelcontextprotocol.io/) (MCP), an open standard that lets AI tools talk to external services securely. Your existing Limio credentials handle authentication, so there is no separate account to create. Claude Desktop connects via [custom connectors](https://support.claude.com/en/articles/11175166-get-started-with-custom-connectors-using-remote-mcp); Claude Code connects via the [MCP CLI](https://code.claude.com/docs/en/mcp).
## How it works
You connect the MCP server to an AI assistant (like Claude Desktop or Claude Code). When you ask the assistant something about your Limio environment, it calls the server on your behalf, using your credentials. The assistant then returns the results in a readable format.
You stay in control. The assistant asks for your approval before taking any action that modifies data.
## Security and data privacy
The MCP server is designed with data isolation and customer control in mind:
* **Bring your own AI key.** The MCP server does not include an AI subscription. You use your own Claude, ChatGPT, or Gemini account. Your contractual relationship with the AI provider is between you and that provider. Limio is not a party to it.
* **No PII exposed.** Order queries use a limited index database, not the primary data store. The data available through MCP includes offer details, order numbers, dates, and aggregate counts, but no customer personal information.
* **Authenticated access only.** Only users with valid Limio credentials can connect. The server uses your existing Limio login, the same credentials you use to access the Limio app.
## Permissions and governance
The MCP server provides several layers of control over what actions can be taken:
* **Per-user access control.** MCP access is **off by default** for every user, existing and new. An admin — anyone with User Management access — turns it on for specific users from the user's profile screen, and the All Users list shows each user's MCP Access state in a read-only column. Turning access off takes effect immediately, and a user without access receives a clear error when their AI tool tries to connect. This lets you start with a small, trusted group and widen access as you build confidence.
* **Connector-level tool restrictions.** In Claude Desktop, you can [configure your connector](https://support.claude.com/en/articles/11503834-building-custom-connectors-via-remote-mcp-servers) to expose only read-only tools (retrieve offers, query orders) or include write tools (create, update). This lets you control what the AI assistant is allowed to do at the connection level.
* **Plan mode.** Claude Desktop and Claude Code support [plan mode](https://code.claude.com/docs/en/interactive-mode), where the assistant shows you exactly what it intends to do before executing. This is useful for bulk operations -- -you can review the full list of changes and approve, reject them, or ask for changes.
* **Environment separation.** Set up separate connectors for each environment (sandbox, pre-production, production). This prevents accidental changes to the wrong environment and lets you test workflows safely before applying them to production.
* **No destructive operations.** No MCP tool can delete data, for any user.
{% hint style="warning" %}
**Current limitation:** The MCP server does not yet enforce Limio app-level role permissions — a user with MCP access enabled can use all enabled tools, regardless of their role within the Limio app. Role-based enforcement is planned for a future release. In the meantime, use the per-user access switch, connector-level tool restrictions and environment separation to manage access.
{% endhint %}
## Capabilities
| Area | What you can do |
| ----------- | ----------------------------------------------------------------------------- |
| **Offers** | Retrieve, create, and update offers, including volume and usage-based pricing |
| **Pages** | Find which pages display a given offer |
| **Orders** | Query order and subscription events with filters |
| **Pricing** | Recurring, one-time, volume-tiered, and usage-based pricing types |
## Next steps
* [Setup](/product/mcp-server/setup.md): connect the server to your AI assistant
* [What You Can Do](/product/mcp-server/usage.md): a guide to available actions
* [Pricing Reference](/product/mcp-server/pricing.md): supported pricing configurations
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:
```
GET https://docs.limio.com/product/mcp-server/mcp-server.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.