AIght_
ToolsLearnFieldsUniverseSignalHumanAbout
Take the quiz
← All concepts

Concept

AI Agents

When AI stops answering and starts doing — and then, very often, hits a wall

Mankaran Singh·Updated May 17, 2026

Where this idea lives

PREREQUISITESTOOLS THAT SHOW ITAI AgentsFunction CallingFunction Calling — The JSON-shaped API that turned chat models into clients of the real world.Prompt EngineeringPrompt Engineering — The craft of talking to a model that will take you exactly as literally as it decides toModel Context ProtocolModel Context Protocol — The open standard that lets AI models talk to your tools without a custom integration per modelMultiagent SystemsMultiagent Systems — What happens when AI models try to coordinate with each other — and the new failure modes that come with itCursorCursorReplit AgentReplit AgentWindsurfWindsurfBolt.newBolt.newCommon misconception: Agents are tireless workers that figure it out eventually.Common misconception: More tools means a smarter agent.Common misconception: An agent that loops is making progress.
prereqsrelatedtoolsmisconceptions
shows up in:Software EngineeringSales & Business DevelopmentMarketing & Advertising
You might think:Agents are tireless workers that figure it out eventually.More tools means a smarter agent.An agent that loops is making progress.

The word "agent" has been stretched so thin it's nearly useless. Every product calls itself agentic now. Assistants, pipelines, chatbots with a search button — all agents, according to their marketing. Let's actually be specific about what the word means when it's used carefully.

A chatbot responds. You ask, it answers. The interaction is complete.

An acts. It takes your goal, figures out the steps, runs them, looks at what came back, adjusts, and keeps going until the goal is reached — or until it decides it can't.

That last part matters too.

If "agent" makes it into the title of a product page in 2026, you have basically zero information about what's inside.

The key distinction isn't capability. It's the loop.

§

The loop is the thing

An agent's defining trait is that it runs in a feedback loop with the world. It doesn't just generate output — it observes state, picks an action, takes it, observes the new state, and repeats. The simplest version of this pattern is sometimes called [·], shorthand for Reason and Act.

Observe what the current state is. What tools are available? What's already happened? What does the environment look like?

Reason about what to do next. Not necessarily deep reasoning — sometimes it's just: "I need to search the web before I can answer this."

Act by calling a tool, writing a file, sending a request, generating text.

Observe the result and decide what's next.

def run_agent(goal: str) -> str:
    messages = [{"role": "user", "content": goal}]

    while True:
        response = llm.complete(messages, tools=AVAILABLE_TOOLS)

        # If the model stopped calling tools, it's done
        if not response.tool_calls:
            return response.content

        # Execute each tool call, feed results back into context
        for call in response.tool_calls:
            result = execute_tool(call.name, call.arguments)
            messages.append({"role": "tool", "content": str(result)})

        messages.append(response)
“Most agents fail not because the AI is wrong — but because the feedback loop is brittle.

The model is often fine. The scaffolding around it is what cracks.


◉ INTERACTIVE

Pick a goal to watch an agent work:

What separates agents from assistants

You can give a chatbot tools and it'll use them. That's not quite the same as an agent.

What makes something agentic is multi-step autonomy — pursuing a goal across multiple actions without a human stepping in at each one. An assistant with will search the web if you ask it to search. An agent will search, read the results, notice they're useless, search again with a refined query, extract the relevant sections, cross-reference them, then write its response.

The agent decides what to do at each step. You set the goal once.

"Autonomy" sounds glamorous. In practice it mostly means "no human to blame when the agent does something stupid."

This is why planning matters so much in agent design. A good agent doesn't just execute the first thing it thinks of — it considers the goal, breaks it into subproblems, and sequences actions in a sensible order. Whether the planning is explicit (the agent writes out a plan first) or implicit (the model just reasons through steps naturally) varies a lot by approach.


The hard parts nobody talks about first

Building an agent is easy. Building one that works reliably is much harder, and almost none of the difficulty lives where you'd expect.

Your agent isn't failing because the AI is dumb. It's failing because the web search timed out and the JSON parser hit a quote it didn't like.

Tool reliability is the ceiling. An agent is only as good as its tools. If a web search returns garbage, or a file reader silently fails, the model is reasoning on bad premises. Agents amplify whatever quality your tools have — for better or worse.[·]

Context accumulates. Each tool call adds more to the conversation. Long agent runs hit token limits, or the model loses track of the original goal under the weight of intermediate results. Good agent design thinks about what to trim, summarize, or move out of .

Error handling needs explicit design. When a human gets a confusing result, they reconsider. An agent will keep going unless you've built in logic to recognize that something went wrong and try a different approach. Production agents invest heavily in retry logic, error pattern detection, and graceful exits.

If your agent runs longer than you'd be comfortable supervising in person, you probably want some "are you sure this is going well?" checkpoints in the loop.

Major providers now ship official tool-use APIs[·][·] that handle a lot of the plumbing for you — schema parsing, message routing, error envelopes. Use them. Hand-rolling this stuff is a rite of passage you'll regret.

§

The most useful mental model: an agent is a model running inside a while loop, with tools as its hands and observations as its eyes. What makes an agent good isn't magic in the model — it's the quality of the loop, the tools, and the scaffolding that keeps the whole thing from going sideways.

The loop is ordinary. Done well, what it can do is not.

← Back to all conceptsBrowse tools →
intermediate
Read time9 min read
UpdatedMay 2026
Sources4

Read next

  1. Model Context Protocol →
  2. Multiagent Systems →
  3. Prompt Engineering →