Skip to content

Connect Docebo to ChatGPT: Manage Courses and Learner Enrollments

Learn how to connect Docebo to ChatGPT using a managed MCP server. Execute LMS workflows, enroll learners, and manage courses directly from chat.

Nachi Raman Nachi Raman · · 9 min read
Connect Docebo to ChatGPT: Manage Courses and Learner Enrollments

If you need to connect Docebo to ChatGPT to automate learner enrollments, manage training materials, or audit compliance reports, you need a Model Context Protocol (MCP) server. This server acts as the translation layer between ChatGPT's function calls and Docebo's extensive REST API. You can either build and maintain this infrastructure yourself, or use a managed integration platform like Truto to dynamically generate a secure, authenticated MCP server URL.

If your team uses Claude, check out our guide on connecting Docebo to Claude or explore our broader architectural overview on connecting Docebo to AI Agents.

Giving a Large Language Model (LLM) read and write access to an enterprise Learning Management System (LMS) like Docebo is a massive engineering challenge. You have to handle complex relational data payloads, navigate fragmented endpoints for courses versus sessions, and deal with asynchronous background jobs for reporting. Every time a developer adds a new custom field to a user profile, your custom server code must be updated, redeployed, and tested.

This guide breaks down exactly how to use Truto to generate a secure, managed MCP server for Docebo, connect it natively to ChatGPT, and execute complex LMS workflows using natural language.

The Engineering Reality of the Docebo API

A custom MCP server is a self-hosted integration layer. While the open MCP standard provides a predictable way for models to discover tools, implementing it against Docebo's highly specialized enterprise API is exceptionally painful.

If you decide to build a custom MCP server for Docebo, you own the entire API lifecycle. Here are the specific integration challenges that break standard CRUD assumptions when working with Docebo:

Fragmented Resource Hierarchies

Docebo's architecture heavily separates learning objects. Courses are distinct from Learning Plans (LPs), which are distinct from ILT (Instructor-Led Training) Sessions and Certifications. When an LLM asks "Enroll John in the Compliance training," your server must understand if "Compliance" is a Course, a Session, or a Learning Plan, as each requires a completely different API endpoint (e.g., /learn/v1/enrollments vs /learningplan/v1/learningplans/{learning_plan_id}/enrollments/bulk). Truto's auto-generated tools map these explicitly, allowing the LLM to search catalogs and pick the correct enrollment route.

Background Jobs for Heavy Operations

Standard REST assumes synchronous requests and responses. However, Docebo relies heavily on background jobs for massive operations - like generating bulk user reports or archiving course enrollments. When calling /report/v1/userreport, the API does not return data; it returns a job hash or ID. Your AI agent must know to poll /report/v1/userreport/{job_id} until the report is ready. If you build this from scratch, you must write stateful polling loops into your MCP server.

Multi-Domain and Org Chart Complexities

Enterprise Docebo instances rely on "Multidomains" (extended enterprise clients) and complex Org Charts (branches). A standard AI tool might easily assign a user to the wrong branch or domain if the structure isn't mapped properly. Managing users requires careful handling of the /manage/v1/orgchart branch IDs, not just flat tenant IDs.

Rate Limits and Normalized Headers

When ChatGPT executes multiple bulk API calls rapidly, you will inevitably hit Docebo's API rate limits.

A critical factual note: Truto does not retry, throttle, or apply backoff on rate limit errors automatically. When the upstream Docebo API returns an HTTP 429 Too Many Requests, Truto passes that exact error back to the caller (your ChatGPT agent). However, Truto heavily normalizes the upstream rate limit information into standardized HTTP headers per the IETF specification (ratelimit-limit, ratelimit-remaining, ratelimit-reset).

The caller (the AI agent framework) is responsible for reading these headers and executing its own retry and backoff logic. Truto does not absorb rate limit errors on your behalf, but it gives you exactly the standardized data you need to handle them gracefully.

Docebo to ChatGPT Quickstart Guide

If you want the fastest path from a fresh Truto account to ChatGPT successfully calling the Docebo API, follow these steps.

What you need:

  • A Truto account with API access.
  • A Docebo administrator account to authorize the connection.
  • A ChatGPT Pro, Plus, Business, Enterprise, or Education seat with Developer mode enabled.

Step 1: Connect Docebo as an Integrated Account

In the Truto dashboard, navigate to Integrated Accounts -> New Integrated Account, select Docebo, and complete the authorization flow. Truto securely stores the refresh token and automatically refreshes access tokens before they expire. ChatGPT never has to handle expired credentials.

Step 2: Create the MCP Server

Truto dynamically generates tools based on Docebo's OpenAPI documentation and your specific configurations. You can generate the MCP Server URL via the UI or the API.

Method A: Via the Truto UI

  1. Navigate to the integrated account page for your Docebo connection.
  2. Click the MCP Servers tab.
  3. Click Create MCP Server.
  4. Select your configuration options (e.g., filter to only read and write methods, or filter by specific tags like courses).
  5. Click save and copy the generated MCP server URL.

Method B: Via the API You can programmatically generate the server endpoint by making a POST request to Truto:

curl -X POST https://api.truto.one/integrated-account/$INTEGRATED_ACCOUNT_ID/mcp \
  -H "Authorization: Bearer $TRUTO_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Docebo for ChatGPT",
    "config": {
      "methods": ["read", "write"],
      "tags": ["courses", "users", "enrollments"]
    }
  }'

The response will contain a url field resembling https://api.truto.one/mcp/<token>. This single URL handles routing, schema provision, and authentication. Treat this URL like a highly sensitive password.

Step 3: Connect the MCP Server to ChatGPT

You can plug this URL into ChatGPT through the standard interface or via a local SSE (Server-Sent Events) proxy.

Method A: Via the ChatGPT UI (Custom Connectors)

  1. Open ChatGPT and navigate to Settings -> Apps -> Advanced settings.
  2. Enable Developer mode.
  3. Under MCP servers / Custom connectors, click to add a new server.
  4. Name: Give it a recognizable name (e.g., "Docebo LMS").
  5. Server URL: Paste the Truto MCP URL generated in Step 2.
  6. Click Save. ChatGPT will immediately connect, perform the handshake, and list the available Docebo tools.

Method B: Via manual config file (SSE Proxy) If you are running a local orchestrator or utilizing a custom agentic framework that connects to ChatGPT's models, you can define the server in an MCP configuration file (mcp.json):

{
  "mcpServers": {
    "docebo-truto": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-sse",
        "--url",
        "https://api.truto.one/mcp/<token>"
      ]
    }
  }
}

Security and Access Control

Exposing an enterprise LMS to an autonomous LLM requires strict governance. Truto MCP servers provide four primary mechanisms to restrict what an AI agent can do:

  • Method Filtering (config.methods): Restrict the server to safe operations. Setting methods: ["read"] ensures ChatGPT can only list and get records, completely preventing it from accidentally deleting courses or altering user profiles.
  • Tag Filtering (config.tags): Scope access to specific functional areas. By passing tags: ["enrollments"], the agent can manage assignments but has no access to billing or platform settings.
  • Additional Authentication (require_api_token_auth): By default, possessing the MCP URL grants access. By setting require_api_token_auth: true, the MCP client must also pass a valid Truto API token in the Authorization header, enforcing a secondary layer of security.
  • Time-to-Live (expires_at): You can set an exact ISO datetime for the MCP server to automatically destruct. This is perfect for granting temporary access to contractors or short-lived AI agent instances.

Hero Tools for Docebo

Truto maps hundreds of Docebo endpoints. Here are the highest-leverage tools your AI agent can call to automate LMS administration.

list_all_docebo_courses

Fetches a complete list of courses available in the platform. This is the foundation for any AI agent tasked with finding content or validating that a specific training module exists before assigning it.

"Get a list of all active courses related to 'Cybersecurity' in Docebo so we can map out our Q3 compliance training requirements."

create_a_docebo_user

Automates employee onboarding by creating a new user record directly in the LMS. The agent can take data from an HR system and format the JSON payload, mapping usernames, passwords, and branch IDs automatically.

"Create a new Docebo user for jane.doe@example.com, set her role to learner, and assign her to the North America Sales branch."

create_a_docebo_enrollment

Enrolls one or multiple users into specific courses. The AI agent can format the required course_ids, user_ids, and level arguments, translating natural language requests into direct API actions.

"Enroll Jane Doe in the 'Information Security 101' and 'Code of Conduct' courses, and set her enrollment level to student."

create_a_docebo_learningplan

Creates a structured learning path by defining a new Learning Plan (LP). Agents can draft descriptions, set the language codes, and initialize the plan before populating it with courses.

"Create a new Learning Plan called 'Q4 Sales Enablement Bootcamp' with a description outlining our new product launch training."

docebo_learningplan_enrollments_list_2

Retrieves a list of users currently enrolled in a specific Learning Plan. Essential for auditing completion status and checking which employees have fallen behind on mandatory training.

"Pull the list of users enrolled in the 'Q4 Sales Enablement Bootcamp' learning plan so I can see who still needs to start."

docebo_report_report_get_data

Extracts the actual rows of data from a custom report configured by a SuperAdmin in Docebo. The AI can pull this analytical data, interpret the results, and generate summaries.

"Fetch the data from custom report ID 42 and summarize the total number of expired certifications for the engineering department."

To view the complete schema details and the hundreds of other available endpoints, visit the Docebo integration page.

Workflows in Action

When you give ChatGPT direct access to Docebo via MCP, it can chain multiple tools together to execute complex operations autonomously. Here is how a few real-world workflows look in practice.

Workflow 1: Autonomous Employee Onboarding

User Prompt:

"We just hired Alex Smith (alex.smith@acmecorp.com). Create a new user account for him in Docebo under the Engineering branch, find the mandatory 'InfoSec Compliance 2026' course, and enroll him immediately."

How the AI Agent Executes It:

  1. Calls create_a_docebo_user passing Alex's email, name, and the mapped branch ID for Engineering.
  2. Calls list_all_docebo_courses with a search filter for "InfoSec Compliance 2026" to retrieve the exact internal course_id.
  3. Calls create_a_docebo_enrollment passing Alex's new user_id and the course_id to finalize the assignment.

Result: The agent replies confirming Alex's account is live and he is enrolled in the required course, saving an IT admin 10 minutes of manual clicking in the Docebo UI.

Workflow 2: Curriculum Orchestration

User Prompt:

"I need to build a training path for new managers. Create a Learning Plan called 'Manager Foundations'. Then, search for our 'Conflict Resolution' and 'Budgeting 101' courses and add them to this plan."

How the AI Agent Executes It:

  1. Calls create_a_docebo_learningplan with the title and description provided, receiving a learning_plan_id in response.
  2. Calls list_all_docebo_courses to find the IDs for both "Conflict Resolution" and "Budgeting 101".
  3. Calls create_a_docebo_learningplan_course to associate those specific course IDs with the newly created Learning Plan.

Result: The agent provisions a complete, multi-course learning plan from scratch, ready for HR to start assigning to new managers.

sequenceDiagram
    participant User
    participant ChatGPT as ChatGPT
    participant TrutoMCP as Truto MCP Server
    participant DoceboAPI as Docebo API

    User->>ChatGPT: "Create a Manager Foundations Learning Plan..."
    ChatGPT->>TrutoMCP: Call create_a_docebo_learningplan
    TrutoMCP->>DoceboAPI: POST /learningplan/v1/learningplans
    DoceboAPI-->>TrutoMCP: Returns learning_plan_id
    TrutoMCP-->>ChatGPT: Result: learning_plan_id
    
    ChatGPT->>TrutoMCP: Call list_all_docebo_courses
    TrutoMCP->>DoceboAPI: GET /course/v1/courses
    DoceboAPI-->>TrutoMCP: Returns course IDs
    TrutoMCP-->>ChatGPT: Result: course_id_1, course_id_2
    
    ChatGPT->>TrutoMCP: Call create_a_docebo_learningplan_course
    TrutoMCP->>DoceboAPI: POST /learningplan/v1/learningplans/{id}/courses
    DoceboAPI-->>TrutoMCP: Success
    TrutoMCP-->>ChatGPT: Result: Courses attached
    ChatGPT-->>User: "The Learning Plan is ready."

Workflow 3: Compliance Auditing and Reassignment

User Prompt:

"Run the 'Expired Certifications' report (ID 105). For any user listed whose certification expired this month, enroll them in the 'Annual Recertification' course."

How the AI Agent Executes It:

  1. Calls docebo_report_report_get_data passing report ID 105 to fetch the custom dataset.
  2. The LLM analyzes the JSON response, filtering users whose expiration date falls within the current month, extracting their user_ids.
  3. Calls list_all_docebo_courses to find the ID for "Annual Recertification".
  4. Calls docebo_enrollment_create_batch to bulk-enroll all identified users into the recertification course.

Result: The agent closes compliance gaps autonomously, interpreting analytical data and taking corrective action without human intervention.

Stop Hardcoding LMS Integrations

Building a reliable, AI-ready integration for Docebo means dealing with complex branch hierarchies, asynchronous report generation, and highly nested JSON schemas. Writing and maintaining that boilerplate code is a drain on engineering resources.

Truto abstracts this away. By normalizing authentication, mapping the entire REST surface dynamically from documentation, and serving it over a standardized JSON-RPC interface, you can give your LLMs full access to Docebo in minutes, not months.

FAQ

How do I connect Docebo to ChatGPT?
You can connect Docebo to ChatGPT by using a Model Context Protocol (MCP) server. Truto auto-generates this server, providing a secure URL that you can plug directly into ChatGPT's custom connectors settings.
Does Truto handle Docebo API rate limits automatically?
No. When Docebo returns an HTTP 429 rate limit error, Truto passes it to your ChatGPT agent alongside standardized IETF rate limit headers. Your AI agent framework is responsible for handling the retry and backoff logic.
Can I prevent ChatGPT from deleting Docebo courses?
Yes. When generating the MCP server in Truto, you can use method filtering to strictly limit the server to `read` operations, ensuring the AI agent cannot modify or delete any LMS records.
What is required to connect the MCP Server to ChatGPT?
You need a ChatGPT Pro, Plus, Business, Enterprise, or Education account with Developer mode enabled to add custom MCP connectors.

More from our Blog