EN
Foundations
Tool calling, planning vs reactive, and the limits youβll hit in prod.
Build Your First AI Agent
β ββWe write the simplest agent that actually works - no magic, no frameworks.
Learn the patternImplementationsHow an Agent Uses Tools (Basics)
β ββIt works with text only.
Learn the patternImplementationsHow to Restrict Tool Access
β ββTool calls are where agents break production: schema drift, retries, side effects, and that same 'oops' admin token. Here's how to survive.
Learn the patternImplementationsHow an agent decides what to do next (Planning vs Reactive)
β ββIt is not given an instruction. It is not told the action order.
Learn the patternImplementationsWhy LLM Agents Can Fail: Technical Limits Explained
β ββUnderstand why LLM agents fail: hallucinations, context limits, tool errors, and the engineering guardrails that make agent behavior reliable.
Learn the patternImplementationsAgent Memory: What It Remembers and Why
β ββWithout this, every new action would be like a first attempt. Sometimes endlessly.
Learn the patternImplementationsWhat an agent is allowed to do (and what it is not)
β ββNot all actions are safe
Learn the patternImplementationsWhen an agent should stop (and who decides)
β ββBecause its job is to complete work. Not decide when enough is enough.
Learn the patternImplementations