agno-workshop

This the sample codes and demo suite for Introduction to LLM Workshop with Agno — a lightweight, model-agnostic framework for reasoning agents, multimodal agents, and agentic workflows.

📦 Installation & Environment Setup

0. Installing uv

on mac and Linux:

curl -LsSf https://astral.sh/uv/install.sh | sh

on windows:

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

or with pip:

pip install uv

1. Clone this repo

git clone https://github.com/dsaad68/agno-workshop.git
cd agno-workshop

2. Install dependencies

uv sync --frozen

3. Set up API keys

• For OpenAI: export OPENAI_API_KEY=sk-...
• For Raindrop Agent: export RAINDROP_ACCESS_TOKEN=...

---

▶️ Running the Demos

This repo includes several demo agents to showcase Agno's capabilities:

1. Barebone Agent

A minimal agent using OpenAI GPT-4.1. Answers general questions.

uv run demo/simple/agent.py

2. Simple Tool Agent

Agent with YFinance tools for financial data, company info, and news.

uv run demo/with_tool/tool.py

3. Raindrop Agent

Agent with a custom toolkit for querying Raindrop bookmarks by date and tag. More about Raindrop.io. Requires RAINDROP_ACCESS_TOKEN. Visit Raindrop.io to get your token.

uv run demo/with_tool/raindrop_agent/agent.py

4. Knowledge Agent

Agent that create a simple RAG over PDFs. Requires Qdrant Vector Database.

1. Setup the Vector Database:

docker run -p 6333:6333 -p 6334:6334 \
  -v $(pwd)/qdrant_storage:/qdrant/storage:z \
  qdrant/qdrant

2. Run the agent:

uv run demo/knowledge/qdrant-demo.py

5. Agent with Reasoning Tool

Agent that uses a reasoning tool to answer questions.

uv run demo/reasoning/reasoning_tool.py

6. Agent with Reasoning LLM

Agent that uses a reasoning LLM to answer questions.

uv run demo/reasoning/reasoning_llm.py

7. Agent with Memory

Agent that uses a memory to answer questions.

uv run demo/memory/simple.py

8. Agent Team

Agent Team that contains two agents:

• Web Agent: Search the web for information
• Finance Agent: Get financial data

They are working together to do a research on a given topic.

uv run demo/team/coordinate.py

---

🖥️ Agent Playground (FastAPI App)

You can launch a playground UI to interact with all agents:

uv run main.py

This starts a FastAPI app with a chat interface for all demo agents.

---

💬 Agent UI (Next.js Frontend)

A modern chat UI for Agno agents is included in agent-ui/. The Agent UI doesn't support Agent Team & Agent Workflow.

Quick Start

npx create-agent-ui@latest
cd agent-ui
npm run dev

• Open http://localhost:3000 to chat with your agents.
• By default, connects to http://localhost:7777 (configurable in the UI).
• See agent-ui/README.md for more.