A modular, open-source framework for building secure, AI-assisted enterprise workflows using Agentic AI, Model Context Protocol (MCP), and modern software engineering practices.
Project Status: 🚧 Early Development (Foundation Phase)
The OpenCampus AI Framework (OCAF) is an educational and reference implementation for designing enterprise-grade AI workflow systems.
Although the first reference implementation focuses on higher education administration (registrar services, admissions, and academic workflows), the architecture is intentionally designed to support a wide range of enterprise domains.
The project emphasizes:
- Modular architecture
- AI-assisted workflow automation
- Human-in-the-loop decision making
- Secure enterprise integration
- Production-quality engineering practices
- Comprehensive documentation
Modern organizations have mature systems of record but often lack intelligent orchestration between them.
This project explores how Large Language Models (LLMs), Model Context Protocol (MCP), APIs, workflow engines, and modern software engineering can be combined to create secure, maintainable, and explainable AI-enabled business applications.
The goal is to demonstrate best practices—not just build another AI chatbot.
The project is currently focused on establishing:
- Repository structure
- Architecture documentation
- Development standards
- Learning materials
- Initial platform foundation
Application development will begin in the next milestone.
- REST API Platform
- AI Service Layer
- MCP Server
- Workflow Engine
- Authentication & Authorization
- Document Processing
- Policy Search (RAG)
- Registrar Reference Application
- Additional Enterprise Examples
apps/ Deployable applications
services/ Business services
shared/ Shared libraries
docs/ Project documentation
learning/ Learning materials
.ai/ AI development toolkit
tests/ Automated tests
infrastructure/ Deployment resources
| Document | Purpose |
|---|---|
| docs/Vision.md | Project vision |
| docs/Roadmap.md | Development roadmap |
| docs/Architecture.md | System architecture |
| docs/Glossary.md | Common terminology |
| learning/ | Structured learning guide |
| docs/adr/ | Architecture Decision Records |
If it's explaining a technology (e.g., FastAPI, Docker, MCP), it belongs in learning/. If it's explaining a project-wide decision or process, it belongs in docs/. If it's explaining a specific folder or service, it belongs in that folder's README.md.
- Python
- FastAPI
- PostgreSQL
- Redis
- Docker
- Model Context Protocol (MCP)
- OpenAI-compatible APIs
- Ollama
- GitHub Actions
- Documentation evolves with the software.
- AI is a replaceable component.
- Security is designed in from the beginning.
- Build incrementally with working software.
- Prefer open standards over vendor lock-in.
The project roadmap is maintained in:
docs/Roadmap.md
Contributions are welcome.
Please see:
CONTRIBUTING.md
This project is under active development. The table below tracks the high-level progress of major components.
| Area | Status | Notes |
|---|---|---|
| Repository Structure | ✅ Complete | Initial project organization established |
| Project Documentation | 🚧 In Progress | Core documentation created; will evolve with the project |
| Learning Materials | 🚧 In Progress | Initial learning roadmap created |
| Architecture | 🚧 In Progress | High-level architecture and vision defined |
| Development Environment | ⏳ Planned | Python, VS Code, Docker, uv, PostgreSQL |
| FastAPI Backend | ⏳ Planned | Initial API application |
| Authentication & Authorization | ⏳ Planned | OAuth2 / OpenID Connect / RBAC |
| Workflow Engine | ⏳ Planned | Generic approval and routing engine |
| AI Service Layer | ⏳ Planned | Multi-provider LLM abstraction |
| Model Context Protocol (MCP) | ⏳ Planned | MCP server and business tools |
| Document Processing | ⏳ Planned | OCR, ingestion, metadata extraction |
| Knowledge & RAG | ⏳ Planned | Enterprise knowledge retrieval |
| Reference Applications | ⏳ Planned | Registrar, Admissions, HR, and additional examples |
| Testing & CI/CD | ⏳ Planned | Automated testing and GitHub Actions |
| Production Deployment | ⏳ Planned | Containerized deployment and monitoring |
- ✅ Complete
- 🚧 In Progress
- ⏳ Planned
⚠️ Under Review- ❌ Deferred
Released under the MIT License.