← Daily AI recipesWork✓ Verified Aug 6, 2026

Run All Your AI Agents on One Screen

A unified dashboard to control multiple AI agents at the same time without switching between tabs

  1. 1.Open the Agent OS website and click «Get Started» or «Launch App» on the homepage to go to the web app.

    If the service is unavailable in your region, use a VPN. Registration is free, no card required.

    Agent OS ↗
  2. 2.Create an account or sign in with Google, then on the dashboard click «Create Agent» in the top menu.

    The free plan lets you run several agents at the same time.

    Agent OS ↗
  3. 3.Set up your first agent: give it a name, select the model to connect (GPT-4, Claude, Gemini, etc.) and paste the system prompt into the «Instructions» field.

    To connect models you will need an API key from the corresponding provider (OpenAI, Anthropic, etc.). Keep your keys safe - they are sensitive data.

    Agent OS ↗
    Prompt
    You are a [role, e.g., marketing copywriter]. Your task: [specific task]. Always respond in [language]. Format your output as [format]. Before answering, ask clarifying questions if the request is ambiguous.
    What this prompt doesEnglish system prompt for setting up the agent. Replace [role] with the role (e.g., marketing copywriter), [specific task] - with a task description, [language] - with the desired response language, [format] - with the output format (e.g., bullet list).
  4. 4.Repeat steps 3 - 4 for every agent you want on hand: for example, one for writing tasks, one for data analysis, one for code.

    Each agent appears as a separate card on the dashboard and can be opened in parallel panels.

    Agent OS ↗
  5. 5.Send a task to one or more agents at the same time by clicking their card and typing your request in the chat. Responses appear side by side - on one screen, with no tab switching.

    You can send the same prompt to several agents at once and compare results from different models.

    Agent OS ↗
    Prompt
    Analyse the following text and return: 1) key summary in 3 sentences, 2) tone of voice, 3) suggested improvements. Text: [paste your text here]
    What this prompt doesPrompt for analyzing text. Replace [paste your text here] with your text. The agent will return a summary, identify the tone, and suggest improvements.
💰 What it costs to followprices as of 2026-08
  • freeAgent OS (platform)Fully free open-source platform installed on your computer, with no limitations.
  • freeOpenRouter (model access)Free tier with access to Llama, Mistral, and Gemini Flash models; paid models by subscription from $4/month.
  • freeGoogle AI Studio (Gemini key)Free access to Gemini Flash with a request limit; Pro models require a paid API key.
  • APIOpenAI API / Anthropic APIPaid API access, pay per use; see current pricing on the pricing pages.
Try it and be the first to check in

AI services change fast - interfaces and free limits may differ from what's described.