AGENT PROFILES

Docebo Agents

Docebo · Specialist agents

Specialist agents for training and onboarding: create courses, support learners and automate learning administration. Requires Docebo Learn.

Typical tasks

Availability in Europe
Added to the catalog on Added Docebo Agents as a specialist training suite; Learn requirement, credits and data region explained.Details
01

What it does

5 features

Docebo · Agent suite for training

Product and access

A combined suite within Docebo Learn, available since 28 October 2026. Primarily suited to organizational learning teams.

Core features

5 features from provider sources · reviewed on 2 November 2026. Availability and limitations follow below.

  • Courses from source material

    Authoring Agent generates structured courses with interactive elements and assessments.

    Source 2
  • In-course learning support

    Tutor Agent answers questions using course content and approved knowledge.

    Source 2
  • Reinforce learning

    Reinforcement Agent provides personalized, spaced practice.

    Source 2
  • Select course offerings

    Marketplace Agent compares content and imports selected courses.

    Source 2
  • Custom workflows with AgentHub

    Administrators configure goals, actions, permissions and human approvals without code.

    Source 1

Usable AI models

Models not stated

The agents product page names no models. The sub-processor list shows AWS Bedrock, Vertex AI and Azure OpenAI for Docebo AI features, but these are hosts and are not tied to the agents. The checked pages do not cover model choice or own keys.

Model families named

None named

Model choice
Not stated
Your own API key
Not stated
Router or custom endpoint
Not stated
Local models
Not stated

The rating covers the generally available main access. Partly documented · Sources accessed on

Docebo: Agents · Docebo: Sub-processors list

02

Does it fit my task?

2 practical tasks · 3 requirements and limits

A good fit if …

  • your company already uses Docebo Learn and wants to use agents there
  • administrators should decide which actions need human approval
  • course outlines, assessments or learning plans should be created from existing documents and knowledge sources

Less suitable if …

  • you want a tool without a learning platform and without AI credit consumption
  • you need public per-item prices, as the pricing page lists none

Editorial assessment based on provider documentation. Not a product test.

Practical tasks

Editorial examples based on documented features. No product tests performed.

01

Prepare an onboarding course

Your input
Create an introductory course with knowledge checks from our approved documents.
Possible result
A course draft for subject-matter approval.
What to check
Check accuracy, freshness and access permissions.
02

Compile learning options

Your input
Compare suitable courses for training our service team.
Possible result
A selection for review and subsequent import.
What to check
Check licensing costs, language, learning goals and audience before importing.
EDITORIAL EXAMPLE

Create an introductory course with knowledge checks from our approved documents.

Your possible prompt

Organize these shared documents by topic. Create an overview with deadlines and next steps. Prepare draft replies for my approval.

Adjust the task, data and permissions.

Requirements and limits

  • Requirement

    Docebo Learn required

    Requires Docebo Learn, administrator setup and approved content and actions.

    Source 2
  • Limitation

    Contract and AI Credits

    Custom platform quote; Docebo Agents consumes AI Credits. Public unit prices are unavailable.

    Source 3
  • Not yet confirmed

    Value in your learning operation

    The suite requires the learning platform. Time-saving claims are provider reports; effort and value in your own learning operation require assessment.

    Source 2
03

Can I use it?

With requirements

Europe & regional availability

Announced for Learn customers; sources state no EU-wide exclusion. Confirm the specific Agents data processing contractually.

Reviewed scope: Access and data processing in Germany and the EU
Reviewed on:

  • With requirements

    With a Learn contract and data review

    The subprocessor list names EU regions for certain services and AI features. Dated January 2026, it provides limited confirmation of data flows for the newly launched Agents suite.

    Region, reason and source

    Region covered: Germany / EU

    Documented reason: Contract and configuration conditions; reviewed sources do not establish a specific restriction caused by the AI Act, DMA or GDPR.

    Official source · www.docebo.inc · Official source · tos.docebo.com

Europe and regulation explained

Accounts & permissions

Based on provider documentation. Permissions depend on the account, integration and organization policies. Review the permissions actually requested before use.

Configuration covered: Docebo Agents within a company's Docebo Learn platform

Sources reviewed:

Show all 3 points
  • Open in reviewed sources

    Account used

    The reviewed pages do not say which account the agents act with. They only state that agents respect the permissions of connected source systems and everyone sees only what they are allowed to see.

    Evidence · Evidence
  • Documented

    Admin approval

    Administrators decide which agents run and who may use them, and which actions need human approval beforehand. They can see and manage AI credit usage; in the plan they set access and limits.

    Evidence · Evidence
  • Documented

    Data access & team sharing

    According to the product page, agents use Docebo learning data, skills profiles and uploaded knowledge as well as connected sources such as Google Drive, SharePoint, Confluence, Slack and Salesforce. The announcement names SharePoint, Confluence, Google Drive and Slack as approved sources; knowledge inherits the permissions of each source.

    Evidence · Evidence
04

Memory and conversation continuity

0 of 6 criteria documented

Learner profiles and earlier practice answers provide context; general memory remains unspecified.

Docebo Learn with Agents, October 2026 launch offering

Sources accessed:

Documentation by criterion

StoragePartly documented

Storage and scope

Agents use learning history, skills profiles, courses and company knowledge. A separate general store for learned conversation facts is not specified.

SelectionPartly documented

Selection and timing

The Reinforcement Agent builds practice on earlier answers. Selection and storage of broader conversation memories are not explained.

Edit & deleteNot publicly established

Inspect, edit and delete

Custom agents' instructions and permitted actions are configurable. A separate memory view with fact editing and deletion is not described.

Access boundariesPartly documented

Agent and channel boundaries

Context concerns learners, courses and authorized company knowledge. Isolation of durable memories between agents or channels remains unspecified.

RetrievalPartly documented

Retrieval

Tutor answers draw on course material and approved knowledge; reinforcement uses earlier answers. This establishes domain context, not a general memory search index.

New conversationsPartly documented

Session reset and compaction

Practice continuity after a course ends is described. Specific chat reset, compaction and long-term memory retention rules are not published.

05

Cost and effort

Prices as of 2 November 2026

Prices and included usage

Pricing checked

Custom platform quote; Docebo Agents consumes AI Credits. Public unit prices are unavailable.

  • The pricing page lists AI Credits as an add-on for Elevate and Enterprise. Confirm allowance and consumption in the quote.

Pricing details

Setup

Requires Docebo Learn, administrator setup and approved content and actions.

Operation & oversight

Based on public provider documentation. Assessments are editorial inferences from the stated properties. Actual time requirements and reliability have not been measured.

Deployment covered: Hosted platform, agents configured through AgentHub

Sources reviewed:

Editorial assessment

Control through approvals, permissions and logs sits with the platform's administrators. Error handling and recovery are not documented and should be clarified with Docebo before use.

Show all 6 points
  • Open in reviewed sources

    Files and working state

    The reviewed pages do not describe what working state agents keep between runs. The announcement explicitly does not cover retention or data residency.

    Evidence
  • Open in reviewed sources

    Sign-ins

    Validity and expiry of connections to source systems are not described in the reviewed sources.

    Evidence
  • Documented

    Starting further runs

    In the no-code builder AgentHub you choose whether an agent runs on a schedule or when conditions change. Agents can monitor conditions, run multi-step workflows, send reminders and generate reports.

    Evidence
  • Documented

    Human intervention

    According to the announcement, administrators decide which actions need human approval before they happen. The product page also mentions escalation rules and gradually expandable autonomy.

    Evidence · Evidence
  • Documented with conditions

    Detecting errors

    The product page promises audit-ready activity logs with traceable agent actions; the announcement mentions a log for IT, security and compliance. The pages do not describe how failures are shown or reported.

    Evidence · Evidence
  • Open in reviewed sources

    Resuming after failures

    Resuming or retrying after failures is not described in the reviewed sources.

    Evidence
06

Evidence and open questions

4 sources · 1 open point

Open points

The reviewed sources contain no reliable information on these points. A gap says nothing about the product.

Requirements and limits: not yet confirmed

Sources and evidence

Provider sources reviewed; over 50 pilot customers according to Docebo. Sources document the offering; no hands-on testing conducted. Example tasks are editorial.

Editorial team: Michael Meierhoff · sources reviewed on . Provider information is not independent evidence of performance.

Read profile as Markdown

How we assess the sources

Provider

Publicly listed provider of an enterprise learning and training platform.

Operator
Docebo Inc.
Organization
Provider
Provider sourcesSource 1 Reviewed
Changes to this profile 1
  • Added Docebo Agents as a specialist training suite; Learn requirement, credits and data region explained.

These notes describe editorial changes. Source and pricing review dates appear in their respective sections.