Administration

Administration

Manage workspaces, users, and platform settings.

---

Workspaces

Overview

Workspaces organize agents, users, and settings for teams or departments.

Create a Workspace

1. Navigate to Administration → Workspaces
2. Click Create Workspace
3. Configure:
   • Name and description
   • Default settings
   • Member access

Workspace Settings

Setting	Description
Name	Display name for the workspace
Description	Purpose and scope
Default model	Default LLM for new agents
Guardrails	Workspace-level safety settings
Branding	Custom logo and colors

---

User Management

User Enrollment

Add users to AI for Work:

1. Go to Administration → User Management → Enrollments
2. Choose enrollment method:
   • Manual — Add individual users
   • Bulk import — Upload CSV
   • Directory sync — Connect to Azure AD, Okta, etc.
3. Assign to workspaces and groups

User Groups

Organize users for access control:

Group: Engineering Team
Members: 45 users
Workspaces:
  - Engineering Workspace (full access)
  - Company Knowledge (read access)
Agents:
  - Code Assistant
  - Documentation Bot

Directory Sync

Sync users from identity providers:

Provider	Features
Azure AD	Users, groups, automatic sync
Okta	SCIM provisioning, groups
Google Workspace	Users, organizational units

Configuration:

1. Go to User Management → Directory
2. Select identity provider
3. Configure connection credentials
4. Map groups to workspaces
5. Enable automatic sync

---

Assist Configuration

Announcements

Display messages to users:

1. Go to Assist Configuration → Announcements
2. Create announcement:
   • Title and message
   • Target users/groups
   • Start and end date
   • Display location

Channel Settings

Configure deployment channels:

Channel	Settings
Teams	App permissions, available agents
Slack	Workspace access, channels
Web	Embed code, styling, domains

Guardrails

Configure safety settings:

• Content filtering — Block inappropriate content
• PII protection — Detect and handle sensitive data
• Topic restrictions — Limit conversation scope
• Response limits — Maximum response length

---

LLM Configuration

Model Settings

Configure available AI models:

Provider	Models
OpenAI	GPT-4, GPT-4o, GPT-3.5-turbo
Anthropic	Claude 3 Opus, Sonnet, Haiku
Azure OpenAI	Deployed models
Custom	Self-hosted or fine-tuned

API Keys

Manage API credentials:

1. Go to LLM Configuration → API Keys
2. Add provider credentials
3. Set usage limits (optional)
4. Enable/disable for workspaces

Default Models

Set default models for:

• New agent creation
• Fallback when preferred model unavailable
• Specific agent types

---

Audit Logging

Activity Logs

Track platform activity:

Event Type	Examples
User actions	Login, agent interaction, feedback
Admin actions	Configuration changes, user management
System events	Deployments, errors, integrations

Log Retention

Configure retention policies:

• Standard: 90 days
• Extended: 1 year
• Compliance: Custom retention

Export

Export logs for compliance:

1. Go to Administration → Audit Logs
2. Set date range and filters
3. Select export format (CSV, JSON)
4. Download or schedule delivery

---

API Access

API Keys

Generate API keys for programmatic access:

1. Go to Administration → APIs
2. Click Create API Key
3. Configure:
   • Name and description
   • Permissions scope
   • Expiration (optional)
4. Copy and securely store the key

Rate Limits

Tier	Requests/min	Requests/day
Standard	60	10,000
Professional	300	100,000
Enterprise	Custom	Custom

---

Related

• Security
• Analytics
• Getting Started