Students with no budget: VS Code, a free assistant and one click | DATASPHERES AI - Dataspheres AI

What you will do Set up the assistant route from the earlier lessons without spending anything: a free editor, a free assistant inside it, a free Dataspher...

What you will do Set up the assistant route from the earlier lessons without spending anything: a free editor, a free assistant inside it, a free Dataspheres account, one click to connect them, and a first conversation that uses the skills. Every free tier below is cited to the publisher's own page with the date it was read, 15 October 2026, because these terms change and the page you open outranks this lesson. Install VS Code Download the editor from code.visualstudio.com/download and run the installer for your operating system. Installing needs no account. Open it once so it completes its own setup. Choose a free assistant GitHub Copilot Free is the assistant this lesson uses. It runs inside VS Code and accepts the one-click Dataspheres install link. Make a GitHub account if you have none, then in VS Code hover over the Copilot icon in the status bar, choose Use AI Features and sign in; an account with no Copilot subscription "will be signed up for the Copilot Free plan" (Visual Studio Code docs, Copilot setup, 15 October 2026). The plan costs $0 and allows 2,000 code completions and 50 chat requests a month, with agent mode and the terminal client included (GitHub, Copilot plans page, 15 October 2026). GitHub Copilot Student is the larger allowance for people who can prove enrolment. "GitHub Copilot Student is available to verified students", with unlimited code completions, an allowance of AI credits for chat and agent use, and models chosen by auto selection (GitHub Education, Student Developer Pack, 15 October 2026). Apply at education.github.com/pack with a school email or a dated student ID. Copilot Free works while the application is reviewed. Google Antigravity CLI is a free terminal agent that can run in the VS Code terminal. It replaced Gemini CLI for free individual accounts on 18 June 2026 (Google, gemini-cli discussion 27274, 19 May 2026), so guides that promise free Gemini CLI quota are out of date. Install it and sign in with a personal Google account; accounts without a paid Google AI plan get "a meaningful quota, refreshed weekly" and unlimited tab completions (Google, Antigravity plans page, 15 October 2026). The one-line installer for macOS, Linux and Windows is on Google's getting-started page ; it places a command named agy on your path, and the first run opens the Google sign-in. OpenAI Codex was checked and is not a free door. OpenAI lists Codex in the terminal and IDE extension for Plus, Pro, Business and Enterprise plans, and describes the free ChatGPT plan only as a way to "explore Codex capabilities on quick coding tasks" (OpenAI, ChatGPT pricing page, 15 October 2026). Leave it out unless a plan is already paid for. Create a free Dataspheres account Sign up at dataspheres.ai with Google, Apple or an email address. The free plan is the one you land on and it asks for no card. Create a datasphere, or accept an invitation to one a classmate or teacher already runs; that is the workspace the assistant will read and write. Click Install in VS Code While signed in, open the Connect tab of the Developer Portal and press Install in VS Code , or use the same link here: Install Dataspheres in VS Code . VS Code asks you to trust the server at https://dataspheres.ai/mcp , then opens a Dataspheres sign-in in your browser. Approve it and you are connected. No key is pasted and nothing is downloaded. If the link does not open VS Code, write the configuration by hand into .vscode/mcp.json in an open folder: { "servers": { "dataspheres": { "type": "http", "url": "https://dataspheres.ai/mcp" } } } Start with the prompts Open Copilot Chat, switch to agent mode, and ask it to list your dataspheres. It calls get_context and answers with each datasphere, your role in it and the owner's brief for that space, which is the same check the earlier lesson described. Then use the skills, which appear under Add Context as /mcp.dataspheres.<skill> . Run pages before asking for a page, house-style before asking for any prose, and planner before asking about tasks. A good first request is a page in your datasphere that summarises this week's reading: the assistant files it in the right folder, writes it in the house voice and returns the link.