AI Agents

What is an AI Agent?

AI Agents are advanced systems using LLMs as their “brain” to independently understand instructions, make decisions, and take actions.

How Agents Differ from Regular Chatbots

• Connect with additional tools, memory systems, and planning abilities
• Capable of working with websites, accessing databases, and completing complex multi-step tasks
• Can take actions in the real world (send emails, schedule meetings, etc.)

The Three-Step Process

AI Agents follow a simple pattern:

1. Notice things: Take in information from their environment
2. Think about it: Consider what to do based on their programming
3. Take action: Do something based on their thinking

Building AI Agents: Key Considerations

Define Purpose:

• What specific tasks should the agent handle?
• Which current processes waste staff time?
• Where would automation make the biggest positive difference?

Document Workflow:

• Identify all decision points, exceptions, and required information
• Create detailed system prompts with clear instructions
• Leverage existing librarian skills: organizing information, analyzing workflows, understanding user needs

Pro Tip

Focus on automating mundane tasks to free up time for more meaningful work.

Video content coming soon – placeholder for video embed