AI Agents

An AI agent is an LLM connected to tools and running in a loop. The LLM decides what to do, calls a tool, observes the result, and repeats until the task is done. Without the loop and tools, it is just a chatbot.

Core Agent Loop

Every agent follows this pattern:

1. OBSERVE  - Receive input (user message or tool result)
2. THINK    - LLM reasons about what to do next
3. ACT      - Call a tool or return a final answer
4. OBSERVE  - Get tool result, go back to step 2

The loop terminates when the LLM decides no more tool calls are needed and returns a final response. A maximum iteration limit prevents runaway loops.

Tool / Function Calling