> For the complete documentation index, see [llms.txt](https://help.supplyd.co/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.supplyd.co/mcp/what-you-can-do/limits-and-expected-behaviour.md).

# Limits and expected behaviour

Understand read-only access, result limits, ambiguity and responsible use.

## Read-only access

Supply'd MCP can retrieve authorised information. It cannot create, change, approve, import, export or delete records.

## Results are intentionally bounded

Large requests may return a limited set of records or a summary. If the answer says the result is incomplete or truncated:

* narrow the date range;
* add a customer, supplier, product, warehouse or status;
* ask for the next smaller group of records; or
* ask for a report instead of every underlying record.

## Similar names require clarification

If several products, customers or suppliers match, Supply'd may return choices. Select the intended record or provide an exact SKU, order number or other reference. This prevents the assistant from silently choosing the wrong record.

## Follow-up questions

A compatible client should preserve returned record references for follow-up questions. If “When was this last produced?” cannot identify the earlier product, repeat the product name or SKU:

> When was product ABC-123 last produced?

## Dates and current state

Use explicit dates when possible. “This week” and “today” are interpreted using the active business and client context, but an exact range is safer for financial comparisons. Inventory questions normally describe current available data unless you ask for activity over a period.

## Access and availability

An empty result can mean there are no matching records, or that the selected connection access and your Supply'd permissions do not cover them. Check both before assuming the data does not exist.

Rate limits, usage allowances and client context limits may also affect unusually large or frequent requests.

## Use AI answers responsibly

AI-generated summaries can omit context or make mistakes. Check important financial, food-safety, production, inventory and operational decisions against the source records in Supply'd.

## Calls, response time and cost

A question may require several calls, for example product search followed by recipe, production and inventory reads. Each server call uses **0.01 Supply'd usage tokens**. Response time also includes the AI client's processing and any additional retrievals.

Ask focused questions and provide exact references to avoid unnecessary searches. Do not assume that a summary covers every record when the response indicates a result limit.

## Client capability updates

Some clients retain a saved tool catalogue. After a Supply'd update, an administrator may need to refresh or rescan the tools and approve the updated definitions. Reconnect only if refreshing does not resolve the issue.


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://help.supplyd.co/mcp/what-you-can-do/limits-and-expected-behaviour.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.