An Enterprise-Grade AI Agent & Knowledge Base Platform

Build, deploy, and manage intelligent AI agents with advanced knowledge retrieval and workflow automation.

Official Website · 简体中文 · Features · Quick Start · Architecture · Documentation

• Why Clouisle
• Features
• Quick Start
• Architecture
• Configuration
• Use Cases
• Roadmap
• Documentation
• Contributing
• License

Modern enterprises face a common challenge: data fragmentation, low reusability, and zero intelligence execution. Knowledge is scattered across documents, databases, wikis, and internal tools — but when decisions need to be made, that knowledge remains static and non-actionable.

Clouisle transforms this reality by providing:

• Intelligent Knowledge Management: Not just storage, but understanding and reasoning over your data
• Agent-Native Architecture: AI agents that retrieve, reason, and execute — not just answer questions
• Enterprise-Grade Security: Multi-tenancy, RBAC, SSO, audit logging, and compliance-ready features
• Flexible Integration: Support for 15+ LLM providers and extensible tool system

Think of Clouisle as a living intelligence layer that evolves with your business.

• Multi-Model Support: Configure agents with different LLM providers and parameters
• RAG Integration: Three retrieval modes — disabled, citation, and rewrite
• Knowledge Base Binding: Connect agents to specific knowledge bases for context-aware responses
• Tool Integration: Extend agent capabilities with built-in and custom tools
• Visibility Control: Private, team, or public access levels
• Conversation Management: Multi-turn conversations with history, versioning, and token tracking

• No-Code Interface: Drag-and-drop workflow creation
• 15+ Node Types: Including LLM, Condition, Code Execution, HTTP Request, Tool, Sub-workflow, and more
• Execution Modes: Manual, scheduled, webhook, or API triggers
• Real-Time Monitoring: Streaming execution with live status updates
• Debug Mode: Test workflows before deployment

• Multi-Format Support: PDF, DOCX, XLSX, and more via MarkItDown
• Intelligent Chunking: Configurable chunking strategies with preview and editing
• Vector Search: Qdrant-powered similarity search with configurable thresholds
• Async Processing: Background document processing via Celery

Supports 15+ providers out of the box:

• Multi-Tenancy: Team-based resource isolation and management
• RBAC: Granular permission system with custom roles
• SSO: OIDC, OAuth2, SAML 2.0, and CAS support
• Audit Logging: Comprehensive action tracking with before/after snapshots
• Notification System: In-app, email, DingTalk, WeChat Work, Feishu, Slack, and webhook channels
• API Key Management: Scoped access with expiration and usage tracking

• Built-in Tools: Time/Date, Calculator, Web Search (Tavily), File Parser
• Custom Tools: HTTP API tools with authentication support
• MCP Integration: Model Context Protocol for advanced tool capabilities
• Sandboxed Execution: Secure code execution environment

• Docker & Docker Compose
• Python 3.13+ with uv
• Bun 1.0+

# Start PostgreSQL, Redis, and Qdrant
docker-compose -f deploy/docker-compose.dev.yml up -d

# Copy environment file
cp .env.example .env

# Generate secure passwords and update .env
# IMPORTANT: Set strong random values for these fields:
#   - SECRET_KEY
#   - POSTGRES_PASSWORD
#   - REDIS_PASSWORD
#   - QDRANT_API_KEY

# Install dependencies
uv sync --project backend

# Start the API server (database will be auto-initialized on first run)
uv run --project backend main.py server

# In separate terminals, start Celery workers
uv run --project backend main.py worker
uv run --project backend main.py beat

cd frontend

# Install dependencies
bun install

# Start development server
bun dev

• Frontend: http://localhost:3000
• API Documentation: http://localhost:8000/docs
• Default Admin: Check .env for initial credentials

┌─────────────────────────────────────────────────────────────────┐
│                         Frontend (Next.js 16)                    │
│  ┌──────────┐  ┌──────────┐  ┌──────────┐  ┌──────────────────┐ │
│  │ Dashboard│  │ Platform │  │   Chat   │  │  Auth (SSO/Login)│ │
│  └──────────┘  └──────────┘  └──────────┘  └──────────────────┘ │
└─────────────────────────────────────────────────────────────────┘
                                  │
                                  ▼
┌─────────────────────────────────────────────────────────────────┐
│                      Backend (FastAPI)                           │
│  ┌──────────┐  ┌──────────┐  ┌──────────┐  ┌──────────────────┐ │
│  │  Agents  │  │ Workflows│  │Knowledge │  │  Users & Teams   │ │
│  │  Engine  │  │  Engine  │  │  Bases   │  │    Management    │ │
│  └──────────┘  └──────────┘  └──────────┘  └──────────────────┘ │
│  ┌──────────┐  ┌──────────┐  ┌──────────┐  ┌──────────────────┐ │
│  │   LLM    │  │   Tool   │  │  Audit   │  │   Notification   │ │
│  │ Adapters │  │  System  │  │ Logging  │  │     Service      │ │
│  └──────────┘  └──────────┘  └──────────┘  └──────────────────┘ │
└─────────────────────────────────────────────────────────────────┘
                                  │
                    ┌─────────────┼─────────────┐
                    ▼             ▼             ▼
             ┌──────────┐  ┌──────────┐  ┌──────────┐
             │PostgreSQL│  │  Redis   │  │  Qdrant  │
             │   (DB)   │  │ (Cache)  │  │ (Vector) │
             └──────────┘  └──────────┘  └──────────┘

Backend

• Framework: FastAPI (Python 3.13)
• ORM: Tortoise ORM with AsyncPG
• Task Queue: Celery + Redis
• Vector DB: Qdrant
• LLM Framework: LangChain + LangGraph

Frontend

• Framework: Next.js 16 (App Router)
• Runtime: Bun
• UI: shadcn/ui + Tailwind CSS
• Language: TypeScript

Key configuration options (see .env.example for full list):

Configure via the admin dashboard:

• General: Site name, description, branding
• Security: Password policies, session timeout, login limits
• Registration: Enable/disable, require approval, email verification
• Email: SMTP configuration for notifications
• SSO: Configure identity providers
• Notifications: Auto-notification rules and channels

• User and operator docs: docs/guide/README.md
• Developer and architecture docs: docs/dev/README.md

• Multi-provider LLM support (15+ providers)
• Visual workflow builder
• Knowledge base with RAG
• Enterprise SSO (OIDC, SAML, OAuth2)
• Multi-channel notifications
• Comprehensive audit logging
• Industry-specific agent templates
• Advanced analytics dashboard
• Plugin marketplace
• Mobile application

We welcome contributions! Please see our Contributing Guide for details.

Backend

uv run ruff check .          # Lint
uv run ruff format .         # Format
uv run mypy app/             # Type check
uv run pytest                # Test
uv run python scripts/check_licenses.py  # Dependency license compliance

Frontend

bun run lint                 # Lint
bun run build                # Build
bun run license:check        # Dependency license compliance

Clouisle is open-sourced under the GPL v3 license.

Built with these amazing open-source projects:

• FastAPI - Modern Python web framework
• Next.js - React framework for production
• LangChain - LLM application framework
• Qdrant - Vector similarity search engine
• shadcn/ui - UI component library

Star us on GitHub to support the project
PRs are welcome · Build the future of enterprise AI together