AI Agent
An AI agent is an LLM-driven system that pursues a goal in a loop — calling tools, observing results, iterating — instead of returning one answer.
An AI agent is a system that uses a language model to pursue a goal autonomously: it decides on an action, executes it through a tool, observes the result, and repeats — a loop, not a single answer.
The loop is the whole distinction. A plain LLM call maps input to output and stops; an agent closes the feedback cycle — run the test, read the failure, edit the code, run it again. That makes agents capable of multi-step work (and of recovering from their own mistakes), and it makes their quality depend on more than the model: tool design, memory, and termination conditions matter as much as raw intelligence.
In practice "agent" spans a spectrum of autonomy — from a function-calling loop with three tools, through coding agents like Claude Code, to multi-agent systems with planners and workers. Frameworks like LangGraph and CrewAI (compared here) supply the orchestration scaffolding; the Model Context Protocol standardizes how agents reach tools.
Frequently asked questions
- What makes something an agent rather than a chatbot?
- The loop. A chatbot maps one input to one output. An agent decides on an action, executes it through a tool, reads the result, and decides again — repeating until the goal is met. Tool use, statefulness across steps, and self-correction against observed results are the defining traits.
- What are the core components of an AI agent?
- A model (the reasoning engine), tools (the actions it can take — search, code execution, APIs), an execution loop that feeds tool results back as observations, and usually memory (context within a task, sometimes persistent across tasks) plus guardrails bounding what it may do.
Related
- Agentic AIAgentic AI is the class of AI systems that act toward goals — planning, calling tools, and iterating on results — rather than only generating content.
- Which Agent Framework in 2026? LangGraph vs CrewAI vs AutoGen vs OpenAI Agents SDK vs Claude Agent SDKA decision guide to the major AI agent frameworks — control vs. abstraction, multi-agent models, state and durability, and which fits your project.
- Function Calling (Tool Calling)Function calling lets an LLM request structured invocations of your code: describe tools with schemas, the model emits typed calls, your app executes them.
- Agent MemoryAgent memory is how an AI agent retains information beyond its context window — working state during a task and persistent knowledge across sessions.
- What Is Claude Code?A grounded explanation of Claude Code: an agentic command-line coding tool that reads files, runs commands, and works in a loop toward a goal.
- Production Tool & Function Calling: Feed Errors Back as ObservationsHow agents use tools — the call/observe/retry loop, why errors must return to the model, and the schemas, idempotency, and limits that keep it reliable.
- How Computer-Use Agents WorkInside the perception-action loop that lets AI operate real software — screenshots in, clicks out — plus grounding, reliability, and when to use APIs instead.
- Agent HarnessAn agent harness is the system around the model that makes it an agent — the loop, tools, context management, permissions, and recovery machinery.
- Computer UseComputer use is an AI agent operating software through its real interface — reading the screen, moving the cursor, clicking, and typing like a person would.
- MCP (Model Context Protocol)MCP is the open standard for connecting AI models to external tools and data: write one server, and any MCP client — Claude Code, IDEs, agents — can use it.
- Vibe CodingVibe coding is building software by describing intent in natural language and letting an AI agent write the code, judging results by behavior.