📖Reference: AI Employees

Introduction

This reference guide provides external users with comprehensive technical documentation about AI Employees in Motion. It explains how AI Employees operate, the permissions and boundaries that govern their behavior, how they connect to external services, and what you can expect from them as agents in your workspace.

By reading this article, you'll understand:

• Foundations — What AI Employees are, the terminology used to describe them, and how they operate in manual vs. autonomous modes.

• Permission & Access Control — Whose context an AI Employee runs under, how they inherit workspace permissions, and what data they can access.

• Operating Rules & System Boundaries — The system rulebook that defines their scope, modes of operation, write actions, approvals, and auditability.

• Connections & Integrations — How AI Employees connect to external services, how connections are shared, and what integrations are supported.

• Limitations & Restricted Operations — What AI Employees cannot do, including permission constraints, workspace management restrictions, and data format limitations.

• Error Handling & Failure Modes — How AI Employees behave when things go wrong, including graceful failure, logging, and user notification.

• Security Best Practices — How to use AI Employees safely, including principle of least privilege, audit monitoring, and connection management.

• Lifecycle & Support — How AI Employees are versioned, maintained, and supported over time, plus troubleshooting for common issues.

Who This Is For

This article is written for Motion users who:

• Are setting up or managing AI Employees in their workspace.

• Need to understand how AI Employees handle permissions, data access, and external integrations.

• Want to know what AI Employees can and cannot do.

• Are troubleshooting issues or configuring skills.

• Need to understand the security and compliance implications of using AI Employees.

Note: This is a technical reference, not a how-to guide. For step-by-step instructions on adding or configuring AI Employees, refer to AI Employee Tutorial or tutorial videos.

Quick Navigation

Use the table of contents below to jump to the section most relevant to your needs:

• New to AI Employees? Start with Foundations to understand core concepts.

• Setting up permissions? Go to Permission & Access Control and Default Behavior.

• Connecting external services? See Connections & Integrations.

• Understanding constraints? Read Limitations & Restricted Operations and Error Handling & Failure Modes.

• Troubleshooting issues? Jump to Lifecycle & Support → Troubleshooting.

• Security concerns? See Security Best Practices.

Key Principles

Before diving into the details, keep these core principles in mind:

1. AI Employees are team-level users: They operate as workspace-level agents, not individual user accounts. They inherit workspace permissions just like human team members.

2. Context determines access: AI Employees run under the permissions of either the user who triggered them (manual mode) or the user who set up the trigger (autonomous mode). They cannot access data beyond what that user can access.

3. Permissions are never escalated: AI Employees cannot bypass workspace RBAC (role-based access control). If a human user cannot perform an action, neither can an AI Employee.

4. Connections are account-level: Integrations are managed at the account level and can be shared flexibly with specific users, workspaces, or AI Employees. They are not workspace-scoped.

5. All actions are logged: Every action an AI Employee takes is recorded for compliance, debugging, and security auditing.

6. Approvals happen at setup, not execution: In autonomous mode (skills), approvals occur during skill configuration, not each time the skill runs. In manual mode (chat), approvals happen before each action.