Understanding MCP (connected tools) in eLLM

By Paul Flanders ·

Learn how MCP (Model Context Protocol) connects external tools to eLLM, enhancing the AI assistant's capabilities. Discover the differences between organisation and personal tools, approval processes, and how tool actions are recorded.

What this product is for

An AI assistant becomes far more useful when it can take actions (looking something up in another system, reading a file, or fetching information) instead of only writing answers. MCP (short for Model Context Protocol) is the common way these tools are connected to eLLM. It lets the assistant use approved tools in a controlled, recorded way.

Main features

  • Connect external tools the assistant can call during a conversation.
  • Organisation tools set up by administrators and shared with groups.
  • Personal tools that an individual user connects for their own private use.
  • Approval controls so sensitive actions can require a human to sign off.
  • A full record of every tool action in the audit log.

How it works

What a tool is

A tool is a specific action a connected service offers, for example listing files in a folder, reading a file, or searching. When you ask a question that would benefit from a tool, the assistant can choose to use it and bring the result into its answer.

Organisation tools vs personal tools

  • Organisation tools are connected by administrators on the admin console's MCP page. They are shared with the groups the administrator chooses, and can be set to require approval before they run.
  • Personal tools are connected by an individual user in their settings (if they have the right permission). They are private to that user, available only to them, and run automatically because the user owns the connection.

Approvals and safety

Each organisation tool service has an auto-execute setting. When it is off, the assistant's request to use that tool waits in the approvals queue for an administrator to approve or reject. When it is on, the tool runs immediately. Personal tools always run automatically for their owner. Whichever path is used, every tool action is recorded in the audit log.

How to use it

As an everyday user

  • If your administrator has connected organisation tools for your group, the assistant may use them automatically when relevant.
  • If you have permission, you can connect your own personal tools in your settings; see the Settings article.
  • Where a skill or instruction asks the assistant to use a named tool, it will act on that.

As an administrator

  • Connect a tool service on the MCP page, discover its tools, and assign group access.
  • Decide whether its tools run automatically or require approval.
  • Review tool actions in the approvals queue and the audit log. See the MCP services and Approvals articles.

Things to know

  • Tools you can use depend on what has been connected and which groups you belong to.
  • Personal tools are private to you and run without approval; organisation tools may require approval depending on their setting.
  • Tool service names are unique across the whole system.
  • Every tool action is recorded, so there is always an account of what was done.

Troubleshooting

  • The assistant didn't use a tool I expected: the tool may not be connected for your group, or the question may not have called for it.
  • My request is waiting: an organisation tool with approval turned on holds the action until an administrator approves it.
  • I can't add my own tool: connecting personal tools needs a permission your group may not have. Ask your administrator.

Frequently asked questions

What does MCP stand for?

Model Context Protocol, the standard eLLM uses to connect tools to the assistant.

Do tools always run automatically?

Personal tools do. Organisation tools run automatically only if an administrator has turned auto-execute on; otherwise they wait for approval.

Can other people use my personal tool?

No. Personal tools are private to the user who connected them.

Is tool use recorded?

Yes. Every tool action appears in the audit log.

Summary

MCP lets eLLM connect external tools so the assistant can take useful actions, not just answer from documents. Administrators connect shared organisation tools with group access and approval controls, while permitted users can connect private personal tools. Every action is recorded, keeping tool use safe and accountable.

Need assistance navigating the complexities of eLLM? EssingtonITS offers expert guidance and tailored IT solutions to help you succeed. Visit EssingtonITS.co.uk for comprehensive support.

person people found this useful.

Related

Knowledge base 10 Jun 2026

Personal Notebook in eLLM

The personal notebook in eLLM allows users to organise and store notes, images, and files in a private, searchable timeline. It ensures privacy by keeping contents visible only to the user, and it can be integrated with the assistant for personalised answers.

Knowledge base 9 Jun 2026

Git Commands Quick Reference Guide

This quick reference guide covers essential Git commands for repository setup, staging, branching, merging, and working with remote repositories, making it a handy tool for developers to streamline their workflow.

Knowledge base 9 Jun 2026

Organisation Skills in the eLLM Admin Console

The article explains how the eLLM Admin Console allows organisations to create and manage shared instruction sets called "organisation skills" to ensure consistent responses across teams. These skills can be customised, restricted to specific groups, and integrated with extern…

Expand 29 Jul 2026

How AI can support pupils with SEND

Explore how AI tools are aiding pupils with special educational needs and disabilities by simplifying text, breaking down tasks, and enhancing accessibility. Learn about the considerations schools should make before implementation.

Expand 29 Jul 2026

AI policy and practice in further education

Further education colleges need distinct AI policies due to their unique mix of vocational courses, diverse age groups, and specific funding and inspection requirements. This guidance highlights the importance of tailored AI approaches in FE settings.