Why Cloosphere?
Enterprise-grade Security
Operates within private network environments. Role-based access control (RBAC) and complete audit logs satisfy enterprise security policies.
Maximum Productivity
Custom AI Agents, Knowledge Base integration, and Tool integrations deliver an AI experience optimized for each department’s work.
Cost Efficiency
Per-department/per-user usage tracking and multi-model support let you select the optimal model for each task.
Key Features
AI Chat and Model Management
AI Chat
Natural conversations with various LLMs including Azure OpenAI and Ollama. Attach files, and use extensions like web search, image generation, and code execution.
Model Selection
Choose the right model for the job — GPT-4o, Claude, Ollama local models, and more.
Workspace
Agent
Build custom AI assistants combining a System Prompt, Knowledge Bases, Tools, and Guardrails.
Agent Flow
Design and run multi-agent workflows in a visual builder.
Knowledge Base (RAG)
Upload internal documents and let the AI ground its answers on them. Supports Retrieval-Augmented Generation backed by vector search.
Database (NL-to-SQL)
Query databases in natural language. Supports 10 DBs including PostgreSQL, MySQL, MSSQL.
Glossary
Register domain-specific terminology so the AI uses the correct terms in its answers.
Knowledge Graph
Connect glossaries, databases, and documents into a single graph to deepen agent understanding.
Guardrails
Safely control AI inputs and outputs with PII detection and content filtering.
Prompt Management
Create and share team-wide prompt templates. Invoke them quickly via
/ commands.Skills
Standardize repetitive work with reusable instruction playbooks that agents invoke on their own.
Tool Integration
Extend agent capabilities by connecting to external APIs and MCP servers.
Marketplace
A service catalog for adding document generators and Google Workspace/Microsoft 365 integrations to your agents.
Tags
Tag workspace resources to classify and filter them.
Collaboration, Automation, and Administration
Projects
Manage documents and chats in one space and share them with teammates.
Channels
Real-time team messaging — share AI responses and collaborate with threads and reactions.
Schedules
Run agents, flows, and models automatically at set times with the cron-based scheduler.
Admin Panel
Provides various system setting tabs: user/organization management, model setup, document processing, branding, and more.
Monitoring
Track operations through usage, audit logs, LLM tracing, and guardrail logs. (Admin > Monitoring)
Evaluation
Score LLM response quality automatically/manually and track performance over time. (Admin > Evaluations)
Platform Architecture
Quick Start
1
Sign in
Sign in with SSO (Microsoft Entra ID) or email/password.
2
Get oriented
Familiarize yourself with the main UI layout — sidebar, model selector, chat area, and input box.
3
Start your first chat
Pick a model and start your first AI conversation.
Supported Browsers
Next Steps
Sign In
Connect via SSO or email
Start Your First Chat
Have your first conversation with AI