# IBM i MCP Agents
AI agents for IBM i system administration and monitoring using Model Context Protocol (MCP) tools. This directory contains multiple agent framework implementations, deployment infrastructure, and web interfaces for interacting with intelligent IBM i system agents.
## Directory Structure
```
agents/ # Root directory for IBM i MCP agents
├── agent-ui # Next.js web interface for agents
├── aws # AWS deployment infrastructure
│ └── ibmi-agent-infra-aws
├── docker # Docker-based deployment infrastructure
│ └── ibmi-agent-infra
├── frameworks # Different agent framework implementations
│ ├── agno
│ └── langchain
└── packages # Reusable packages and SDKs
└── ibmi-agent-sdk
```
## Getting Started
### Prerequisites
- **Python 3.12+** (for agent frameworks)
- **uv** (Python package manager)
- **Node.js 20+** (for Agent UI)
- **MCP Server** running in HTTP mode
## Frameworks
The `frameworks/` directory provides different agent framework implementations, allowing you to choose the best solution for your use case.
### Current Frameworks
| Framework | Language | Description |
|-----------|----------|-------------|
| [Agno](./frameworks/agno) | Python | Agents built with Agno's AgentOS
| [LangChain](./frameworks/langchain) | Python | Agents built with LangChain framework |
### Agno (AgentOS)
The Agno implementation uses [Agno's AgentOS framework](https://agno.com) to provide intelligent agents that connect to IBM i MCP tools via HTTP transport.
**Features:**
- Multi-agent orchestration with specialized agents
- Built-in evaluation framework for testing reliability
- Tool filtering for focused agent capabilities
- WatsonX AI model integration support
- HTTP-based MCP transport for distributed deployments
**Quick Start:**
```bash
cd frameworks/agno
# Set your OpenAI API key
export OPENAI_API_KEY=your_openai_api_key_here
# Run multi-agent system
uv run ibmi_agentos.py
```
**Documentation:** See [frameworks/agno/README.md](frameworks/agno/README.md)
### LangChain
**Status:** Planned
Future implementation using [LangChain](https://langchain.com) framework for agent development. This will provide:
- LangChain's extensive ecosystem integration
- Alternative agent orchestration patterns
- Additional model provider options
- LangChain-specific tooling and utilities
## Agent UI
A modern Next.js web interface for interacting with agent frameworks.
**Features:**
- Real-time chat interface with agents
- AgentOS endpoint configuration
- Markdown rendering for agent responses
- Responsive design for desktop and mobile
**Quick Start:**
```bash
cd agent-ui
pnpm install # or npm install
pnpm dev # or npm run dev
# Open http://localhost:3000
```
**Documentation:** See [agent-ui/README.md](agent-ui/README.md)
## Deployment Infrastructure
Deploy IBM i Agno agents using either Docker-based or AWS-based infrastructure.
> More info: [Agno Deployment Documentation](https://docs.agno.com/deploy/introduction)
| Option | Description |
|---------|-------------|
| [Docker](./docker/ibmi-agent-infra) | Containerized agent deployments using Docker Compose |
| [AWS](./aws/ibmi-agent-infra-aws) | Cloud-based deployments using AWS infrastructure (in progress) |
### Docker
Docker-based infrastructure for containerized agent deployments.
- **Location:** `docker/ibmi-agent-infra/`
- **Documentation:** See [README](docker/ibmi-agent-infra/README.md)
**Features:**
- Containerized agent runtime
- Docker Compose configurations
- Isolated agent environments
- Easy local development setup
### AWS (In Progress)
AWS infrastructure for cloud-based agent deployments.
- **Location:** `aws/ibmi-agent-infra-aws/`
- **Documentation:** See [README](aws/ibmi-agent-infra-aws/README.md)
**Features:**
- AWS deployment configurations
- Scalable cloud infrastructure
- Production-ready deployment patterns
## Agent Capabilities
All agent frameworks connect to the IBM i MCP server and can perform tasks such as:
- **System Performance Monitoring**
- CPU and memory utilization analysis
- Job queue monitoring
- Resource bottleneck identification
- **System Administration**
- Active job management
- System configuration queries
- Service status checks
- **Database Operations**
- SQL query execution
- Table and schema exploration
- Data retrieval and analysis
- **Troubleshooting**
- Error message analysis
- System health checks
- Automated diagnostics
## Resources
- **Agno Documentation:** https://docs.agno.com
- **LangChain Documentation:** https://docs.langchain.com
- **Model Context Protocol:** https://modelcontextprotocol.io
- **IBM i MCP Server:** [../README.md](../README.md)
For general IBM i MCP server issues, see the main project documentation.
