What it is
Hello Agents is an educational project about building ИИ agents from the ground up. It covers basic ideas: what an agent is, how it plans, uses tools, stores memory, and interacts with an environment.
It is useful at a time when the word “agent” is broad and overloaded. The project breaks the topic into understandable parts.
How learning is structured
The material is organized as a book and practice route, with online reading, local reading, fundamentals, tools, memory, planning, and applications.
The format moves from terminology to implementation: understand the model, inspect examples, and build a view of when agents are actually useful.
Learning map
This example shows the shape of the learning route: basics, tool use, memory, and evaluation.
## Agent basics
- role and goal
- environment
- actions
## Tool use
- function calls
- external APIs
- safety checks
## Memory and planning
- short context
- long-term notes
- task decompositionWhat is inside
The repository contains chapters, community material, links to the online version, PDF downloads, and future plans.
Much of the material is in Chinese, but the structure is clear: a course that brings theory and practice together.
Strengths
The main strength is system thinking. Instead of random articles, it treats agent systems as a topic with concepts, limits, and engineering choices.
The educational format also works for study groups or as a topic map for self-study.
Limits
The material needs critical reading. Tools and examples in this field age faster than the core ideas.
Educational demos should not be moved directly into production. Real agents need access control, logging, action checks, and quality evaluation.