Category
Giving AI the ability to act
A language model on its own can only write text. The agent layer turns that text into action: models that call tools, follow multi-step workflows, coordinate with other agents, remember what happened last time, and operate real software. These entries explain the agent loop, the protocols that connect models to the outside world, and the guardrails and human checkpoints that keep autonomy honest.
Entries in this category
- Agent frameworks Libraries and SDKs that handle the agent loop, tool wiring, and state so you build behavior instead of plumbing. Read the entry
- Agent memory Mechanisms that let an AI agent carry information across steps and sessions beyond what fits in the context window. Read the entry
- Agentic workflow A multi-step process that chains, branches, or loops model calls so an AI system can complete tasks too big for one prompt. Read the entry
- AI agent A system where a language model plans, calls tools, and loops on results to finish a task with minimal supervision. Read the entry
- Computer use An agent capability where the model reads the screen and issues clicks and keystrokes to operate software built for humans. Read the entry
- Guardrails Programmatic checks on an AI system's inputs, outputs, and actions that block or correct behavior outside defined limits. Read the entry
- Human in the loop A design pattern where a person reviews, approves, or corrects an AI system's actions at defined checkpoints. Read the entry
- MCP (Model Context Protocol) An open protocol that lets AI applications connect to external tools and data through a standard client-server interface. Read the entry
- Multi-agent systems An architecture where several AI agents split a task, each handling one role, coordinated by an orchestrator or a shared plan. Read the entry
- Tool use (function calling) Letting a language model request actions by emitting structured calls that your code executes and whose results feed back in. Read the entry
Where this layer fits
Agents sit on top of everything else in the stack. They are built from core concepts like context windows and inference, they reach into the retrieval layer when they need your data, and because they act on their own, evals are how you find out whether they can be trusted.
Not sure where to start? AI agents is the entry the rest of this category builds on. Or browse everything at once in the A-Z index.