What can I ask an AI agent using the LRS MCP?

What can I ask an AI agent using the LRS MCP?

Think of an AI agent with the MCP as a conversational interface to your LRS. Unlike the Query Builder or Chart Builder in the LRS, you can ask vague and general questions to an AI agent configured with the LRS MCP. However, if you use terms carefully or apply contextual clues, then the AI agent can extrapolate meaning: e.g., an xAPI “actor” is a learner.

LRS MCP Tools

Our LRS MCP includes 79 functions, or “tools”, which an AI agent can use:
  1. Statement Viewer saved report — list, get, create, update, delete
  2. LRS (upstream) — list, get, create, update, delete 
  3. xAPI key — list, get, create, update, delete
  4. API key — list, get, create, update, delete
  5. LRS log — list, get, create, update, delete
  6. Embeddable chart — list, get, create, update, delete
  7. Embeddable dashboard — list, get, create, update, delete
  8. Plugin — list, get, create, update, delete
  9. Custom dashboard — list, get, create, update, delete
  10. Saved script — list, get, create, update, delete
  11. Portal learner — list, get, create, update, delete
  12. Portal content — list, get, create, update, delete
  13. Portal course — list, get, create, update, delete
  14. Portal class — list, get, create, update, delete
  15. Portal launch — list, get, create, update, delete
  16. Documents — clear
  17. Statements — clear, query
  18. VQL — query
The last two query functions are the tools you’ll probably use the most.

Queries

If you simply ask the AI agent about metrics of learners or courses, then it uses the query_statements and vql_query tools. For comprehensive queries of multiple learners or courses, the AI agent may format its answer as a Markdown list, HTML table, or SVG chart.

For example, you might ask an AI agent a general question like…

Analyze the activity of Peggy Lane.

Twenty-seven seconds later, the AI agent replies…

Key Analysis Summary for Peggy Lane

• Total Activity Volume: 148 xAPI statements sent over a span of ~150 days (12 October 2024 to 11 March 2025).

• Consistent Presence: Unlike other actors (who have month-long gaps), Peggy shows steady engagement every month.

• Activity Peaks: Her peaks occurred in January 2025 (38 statements) and March 2025 (36 statements).

• Verb Breakdown: Very balanced lifecycle behavior, led by initialized (18), suspended/resumed/terminated (17 each), and completed (15).

• Primary Focus Areas: Her study was heavily concentrated on: 1. Neural Networks and Deep Learning (27 interactions); 2. Regression Models (22 interactions); and 3. Retrieving, Processing, and Visualizing Data with Python (21 interactions).
 


Although it can work quickly, you should take the time to read-through the method the AI agent used or otherwise confirm its metrics, or both. Use caution when sharing AI-generated information.

In addition to working quickly, the AI agent can build on earlier queries. So, the more that you ask it, the better the AI agent gets at answering. For example, in this instance, the AI agent replied “Unlike other actors (who have month-long gaps), Peggy shows steady engagement…”. This insight happened because the AI agent had scanned all actor’s activity in an earlier query.

Notes
Frequent, complex, or lengthy queries might use a lot of “tokens” or “quota”. If you’re not on a paid “pro” plan for your AI agent, then it might suddenly stop working.

Operations

You can ask an AI agent to change your LRS in some way.

For example, you might ask an AI agent to build a chart in a dashboard in your LRS, for which it uses the update_customDashboard tool …

In my LRS, in the Custom Reports dashboard, build a new bar chart of the Average Sessions per Student for the ten lessons in the Data Science course.

Seven minutes later, the AI agent replies…

I successfully added the Average Sessions per Student bar chart to the Custom Reports dashboard in the LRS. Here's a summary of what I built:

• Chart title: Average Sessions per Student
• Subtitle: Data Science Course — L1 through L10
• Location: Bottom of the Custom Reports dashboard (row 66)
• Width: Full width (10 columns), 4 rows tall

How it works:

1. A $frequentValues stage on object ID groups by lesson URL, counts unique students (actor ID) and unique sessions (context registration) per lesson.

2. An $addFields (average sessions) stage divides sessions by students to get the average.

3. An $addFields (lesson index, lesson name) stage uses $switch to map each lesson URL to its L1–L10 index and short label.

4. A $sort by lesson index stage ensures bars appear in L1→L10 order, not by volume.

5. A $barChart stage renders the bar chart using lesson name as category, average sessions as value.

6. A final $addFields stage sets axis title, tooltip format (2.34 sessions/student), color (#58a6ff), and Y-axis minimum of 1.
 


Notes
Caution! — An update command might overwrite information. So, create a backup, or tell the AI agent to create something new rather than update something existing.




    • Related Articles

    • Where do I install the LRS MCP, and why?

      Some of our users asked, “The AI Integration directions list code generators, not AI agents; why?” Or “Why must I install a code generator to use the MCP?” What the MCP Does A Model Context Protocol (MCP) helps an AI agent work with another system — ...

    • What Types of Charts Can I Make in the LRS?

      In general, you can make charts in a custom dashboard using two methods: using our Chart Builder, or by programming our Veracity Query Language (VQL). Chart Builder Charts The Chart Builder and similar form-based wizards in the Veracity Learning LRS ...

    • How Do I Show Percent of Students?

      In a custom dashboard, you can easily build a bar chart that shows the number of unique students per course. Chart Builder with simulated data offered in the free Veracity Learning LRS. Problem But what if you want to show the number of unique ...

    • How Do I Tag Groups of Students?

      Our Veracity Launch LMS has a Learner Profile Field feature that lets you “tag” or label a student as belonging to a group that you designate. Then, you can use these metadata tags to streamline later operations on those students and groups in the ...

    • How Do I Add ‘Storytelling’ to a Dashboard?

      If you only use the out-of-the-box dashboards, or have a custom dashboard with one widget, then you’re all set. But, if you decide to build a custom dashboard (or more) with many widgets, then you must consider the overall design. “Widgets” include ...