Feat/agentos full app (#40)

* move agent-ui and remove old agent os templates

Signed-off-by: Adam Shedivy <[email protected]>

* feat: implement Tool Call Panel with dynamic display of tool calls

Signed-off-by: Adam Shedivy <[email protected]>

* feat: add streaming tool calls management and in-progress tracking

Signed-off-by: Adam Shedivy <[email protected]>

* enable agent-ui service in Docker Compose configuration

Signed-off-by: Adam Shedivy <[email protected]>

* Refactor IBM i Agent Infrastructure

- Removed the `verify_watsonx_config.py` script as it is no longer needed.
- Introduced `agent_ids.py` to define official agent IDs for consistency across the system.
- Created `config.yaml` to manage agent configurations, including models and settings for various agents.
- Added `config_helper.py` to provide utility functions for applying agent configurations.
- Implemented `config_manager.py` to handle loading and parsing of agent configuration files.
- Developed `config_models.py` to centralize application configuration, loading values from environment variables with validation.

Signed-off-by: Adam Shedivy <[email protected]>

* Refactor agent creation and configuration

- Consolidated agent creation logic into a new `create_ibmi_agent` function in `base_agent.py` to streamline agent instantiation with shared configurations.
- Updated `get_performance_agent`, `get_sysadmin_discovery_agent`, `get_sysadmin_browse_agent`, `get_sysadmin_search_agent` functions to utilize the new factory method.
- Removed redundant parameters and settings from individual agent functions, enhancing maintainability.
- Simplified error messages and improved code readability by reducing line lengths and consolidating similar lines.
- Cleaned up imports and ensured consistent formatting across multiple files.

Signed-off-by: Adam Shedivy <[email protected]>

* fix package versioning and watsonx models

Signed-off-by: Adam Shedivy <[email protected]>

* update config docs

Signed-off-by: Adam Shedivy <[email protected]>

* improve diagram contrast

Signed-off-by: Adam Shedivy <[email protected]>

* update README to rename Docker Compose Stack to AgentOS Stack

Signed-off-by: Adam Shedivy <[email protected]>

* refactor README to streamline agent documentation and improve clarity

Signed-off-by: Adam Shedivy <[email protected]>

* remove whitespace

Signed-off-by: Adam Shedivy <[email protected]>

* update default model config

Signed-off-by: Adam Shedivy <[email protected]>

---------

Signed-off-by: Adam Shedivy <[email protected]>

Author: ajshedivy

agents/README.md

@@ -2,131 +2,6 @@
 
 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:
@@ -142,14 +17,41 @@ All agent frameworks connect to the IBM i MCP server and can perform tasks such
   - 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
+
+## 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
+
+## Available Agent SDKs
+
+The `frameworks/` directory provides different agent SDK implementations, allowing you to choose the best solution for your use case.
+
+| SDK | Language | Status | Documentation |
+|-----|----------|--------|---------------|
+| [Agno](./frameworks/agno) | Python | ✅ Active | [frameworks/agno/README.md](frameworks/agno/README.md) |
+| [LangChain](./frameworks/langchain) | Python | ✅ Active | [frameworks/langchain/README.md](frameworks/langchain/README.md) |
+| [Google ADK](./frameworks/google_adk) | Python | ✅ Active | [frameworks/google_adk/README.md](frameworks/google_adk/README.md) |
+
+## Deployment Infrastructure
+
+Deploy IBM i agents using Agno AgentOS stack with Docker.
+
+**Documentation:** See [docker/ibmi-agent-infra/README.md](docker/ibmi-agent-infra/README.md)  
+
+**What's Included:**
+- **AgentOS API**: RESTful API for agent interactions
+- **IBM i MCP Server**: Automatically configured and running
+- **PostgreSQL Database**: Session and memory persistence
+- **Agent UI**: Optional web interface
+- **Multi-provider LLM Support**: watsonx, OpenAI, Anthropic
 
 
 ## Resources