What this lesson covers
Tools go by many names
A practical, ground-up introduction to AI agents. Understand the ReAct loop, tools, orchestration, memory, and evaluation, then build real agents using Aurora Trading Agent platform as your lab.
Explain the difference between a bare language model, a chatbot, and an AI agent
Learning outcome for this course.
Describe the ReAct loop (Thought → Action → Observation) and why it makes agents agentic
Learning outcome for this course.
Design tool-calling workflows and understand why tools, function calling, and MCP servers are all the same core concept
Learning outcome for this course.
Use autonomy controls (whitelists, approvals) to safely run agents in production
Learning outcome for this course.
Connect memory, scheduling, and subagents into a full autonomous workflow
Learning outcome for this course.
Build traces and evaluators (including LLM judges) to measure agent quality
Learning outcome for this course.
Lesson reading
## Tools Go By Many Names
In the videos, we talked about tools as functions, API calls, JSON objects, CLI commands, and code execution. These are all ways for an AI agent to actually **do** things.
In the real world, you'll hear tools called by many different names depending on the ecosystem:
- **Function calling** - OpenAI's term for having a model generate structured JSON that maps to a function - **Tool use** - Anthropic's term for the same concept - **Skills** - reusable instructions that teach an agent how to perform a task, used by tools like Claude Code and Cursor - **CLI commands** - shell commands an agent can run directly, like `git`, `npm`, or custom scripts - **MCP Servers** - the newest standard, and worth understanding
## What Is MCP?
MCP stands for **Model Context Protocol**. It's an open standard created by Anthropic that defines how AI agents connect to external tools and data sources.
Think of it like USB for AI agents. Before USB, every device had its own proprietary connector. MCP does the same thing for AI tools - it creates one standard protocol so any agent can connect to any tool.
An MCP server is just a service that exposes tools in the MCP format. For example:
- A **Google Calendar MCP server** exposes tools like `create_event`, `list_events`, `delete_event` - A **GitHub MCP server** exposes tools like `create_issue`, `list_prs`, `merge_branch` - **NexusTrade's MCP server** exposes Aurora's tools - create indicators, run backtests, build strategies - so you can use them from Claude Desktop or any MCP-compatible client
## Why This Matters
When you hear someone say "tools," "function calling," "skills," or "MCP servers," they're all talking about the same core concept from the video: **giving the AI a list of things it's allowed to do, with defined inputs and outputs, so it can generate the right parameters and the system can execute the call.**
The name changes. The pattern doesn't.
## Further Reading (Optional)
- [Anthropic's MCP Documentation](https://modelcontextprotocol.io) - the official spec - [NexusTrade MCP Integration](https://nexustrade.io) - Aurora's tools exposed via MCP