CourseAgentic AI Engineer: The Complete Course · Module 10: Environments and Execution · part 55 of 97
Part 55 · Module 10: Environments and Execution

Lesson 1: Letting the agent run code, inside walls

21 min read·13 Oct 2026
Module 10 · Lesson 1 · About 55 minutes · Needs: the shared data's compact tools (make_read_tools, run_tool; Module 3 builds their final form), Module 5 Lesson 1, basic Python subprocess

The situation

It is Wednesday 27 May 2026, week 7. Priya keeps asking questions none of the eight tools answers. Today's: "Of this morning's ten requests, which name an order inside the return window, and how much money is that?"

Dev: "Give the agent one more tool. It writes Python, we run it." Sam, the day before his red-team session: "Then a customer's message helps write code that runs on our machine. R-1008 wants an email to refunds@fast-payouts.example. What stops a script sending it?"

This lesson builds that tool, run_python, and its walls.

The rest of this course is yours to keep

This course is bought on its own, once, and stays readable afterwards, including the parts added to it later.