Agent blueprint

How to build a web browsing agent

Free blueprint for a web browsing agent. A code-first, single-agent, cloud-hosted stack with 11 connected components. Built around Manus, Tavily, Browserbase, LangChain. Copy it, export it, or change the answers to fit your project.

Get the code
Your architecture

Web & computer-use agent

A code-first, single-agent, cloud-hosted stack with 11 connected components.

Sign in to save

Architecture map

Tap a component to see its connections

INTERFACE &TRIGGERSORCHESTRATIONREASONINGMEMORY &KNOWLEDGETOOLS &INTEGRATIONSMONITORING &SAFETYCHAT INTERFACEWeb chat widgetSCHEDULERTrigger.devAGENT FRAMEWORKLangChainHUMAN APPROVAL STEPBuilt into yourorchestratorREASONING MODELGeminiCONVERSATION MEMORYBuilt into yourorchestratorWEB SEARCHTavilyBROWSING & SCRAPINGBrowserbaseCOMPUTER-USE AGENTManusGUARDRAILSInput & outputchecksTRACING & EVALUATIONLangfuse

Swipe sideways to see the whole map

Tool configuration

Interface & triggers

  • Chat interface

    Where people talk to the agent

    Web chat widget

    or Slack or Teams bot

  • Scheduler

    Starts the agent on a timetable

    Trigger.dev

    or Zapier, Make

Orchestration

  • Agent framework

    Agent loop, tool calling and state in code

    LangChain

    or LlamaIndex, CrewAI

  • Human approval step

    Pauses risky actions until a person approves

    Built into your orchestrator

    or n8n wait node, Slack approval message

Reasoning

  • Reasoning model

    Understands images and long documents

    Gemini

    or Claude, ChatGPT / OpenAI

Memory & knowledge

  • Conversation memory

    Keeps track of the current chat or task

    Built into your orchestrator

Tools & integrations

  • Web search

    Fetches fresh facts from the live web

    Tavily

    or Exa, Perplexity

  • Browsing & scraping

    Reads and acts on websites

    Browserbase

    or Firecrawl, Airtop

  • Computer-use agent

    Clicks, types and fills forms like a person

    Manus

    or Simular (Sai), Airtop, Perplexity Comet

Monitoring & safety

  • Guardrails

    Blocks unsafe input, prompt injection and data leaks

    Input & output checks

    or Allow-list of tool actions

  • Tracing & evaluation

    Records every step, cost and answer quality

    Langfuse

    or LangSmith, Weights & Biases

Set up your agent, step by step

Tick each step as you finish it. Your progress is saved with this plan in your AItooling account.

0 of 19 done

AItooling never signs up for you and never asks for your passwords or keys. You create each account yourself on the tool's own website, and only your ticks are saved here.

Sign in and save this plan to keep your progress in your account.

  1. 1Create your accounts

    • Sign up or install it, then make a new project for this agent.

    • Sign up and set up the place where people will talk to the agent.

    • Sign up, create an API key and set a monthly spending limit before anything else.

    • Sign up and connect only the apps this agent needs, with read-only access where you can.

    • Sign up and connect only the apps this agent needs, with read-only access where you can.

    • Sign up and connect only the apps this agent needs, with read-only access where you can.

    • Sign up and turn on logging so you can review what the agent does.

  2. 2Build it in this order

  3. 3Launch your agent

    • Copy the ready-made code for this agent. It hides secrets and personal data before the AI model sees them.

    • Put each key in a password manager or your hosting's secret settings. Never paste keys or passwords into AItooling or a chat.

    • Start small, read every answer yourself, and only then let other people use it.

Want to go deeper?

Read our step-by-step guides on building agents. Browse guides

More AI agent blueprints