Cursor / VS Code

There is no browser sign-in here. These editors configure a remote MCP server with static headers, so you need an Agent Key — minting one is step 1. If you would rather not handle a token at all, Claude Code connects with sign-in and lives in the same terminal.

Setup

1. 1

   Mint an Agent Key

   Open Create → Agents & access, label the key for the machine you are on, and press Mint key. Copy it immediately — it is displayed once and there is no way to read it back afterwards. If you lose it, revoke it and mint another.

   Screenshot coming soon

   The Mint key dialog on TipTap's Agents & access page, showing a freshly minted key with its value obscured.

   /docs/agents/shared/mint-key.png

2. 2

   Add the server to your editor's MCP config

   Replace ttg_agent_YOUR_KEY_HERE with the key you just minted, keeping the Bearer prefix in front of it.

   Cursor — ~/.cursor/mcp.json

   {
     "mcpServers": {
       "tiptap": {
         "type": "http",
         "url": "https://mcp.tiptapgames.com/mcp",
         "headers": {
           "Authorization": "Bearer ttg_agent_YOUR_KEY_HERE"
         }
       }
     }
   }

   VS Code — .vscode/mcp.json

   {
     "servers": {
       "tiptap": {
         "type": "http",
         "url": "https://mcp.tiptapgames.com/mcp",
         "headers": {
           "Authorization": "Bearer ttg_agent_YOUR_KEY_HERE"
         }
       }
     }
   }

   Cursor nests servers under mcpServers; VS Code uses servers. Using the wrong key is the failure worth warning about, because it does not look like one: the config loads, no error appears anywhere, and the server simply has no tools. Copy the block that matches your editor rather than adapting the other one.

   A project-level .cursor/mcp.json works too if you would rather scope it to one repo — but a file holding a key belongs in your ignore list either way.

   Screenshot coming soon

   An editor showing the mcp.json file with the TipTap server configured.

   /docs/agents/cursor/step-1.png

3. 3

   Restart the editor and check the tools are there

   Neither editor reliably picks up a new MCP server without a restart. Once it is back, open the MCP panel and confirm tiptap is listed with its tools enumerated. A server listed with zero tools is the wrong-key symptom from the previous step.

   Screenshot coming soon

   The editor's MCP panel listing the TipTap server as connected with its tools enumerated.

   /docs/agents/cursor/step-2.png

4. 4

   Ask the agent to build

   Open the agent chat and paste this. It creates its own build session and works through to submission.

   Starter prompt

   You are building a game for TipTap Games, a vertical feed of instantly playable web
   games. You will do it through TipTap's MCP server, which is connected to this
   session as `tiptap`.
   
   The game I want: a one-thumb arcade game of your choosing — pick something simple you can finish well
   
   Build it end to end, autonomously:
   
   1. Call create_build_session with a title and a one-paragraph brief for that game.
      Use the session_id it returns for every call after this one.
   2. Call get_game_specification, then get_sdk_documentation. Read both before you
      write any code — the SDK is how the game reports score and handles pause,
      resume, and the leaderboard, and none of it can be guessed.
   3. Call list_categories so you can set a real category at submit. Unknown slugs
      are dropped silently. Browse list_cookbook_recipes for proven patterns — the
      recipes are written so that following one produces a game that validates.
   4. Write ONE self-contained .html file:
      - Inline every byte of CSS and JavaScript. No external scripts, styles, or fonts.
      - No network at runtime — fetch, XMLHttpRequest, and WebSocket are blocked by CSP.
      - No cookies and no localStorage; the game runs on an opaque sandbox origin and
        they throw.
      - Assets as data: or blob: URLs only. 5 MB maximum.
      - Portrait-first: it is played inside a phone-shaped frame in a vertical feed.
   5. Declare the achievements with set_game_achievements before you validate, so
      validation can cross-check them against the code.
   6. Call upload_game_draft with the HTML, then validate_game_draft. Fix every issue
      with severity "error" and validate again — read each issue's suggestedFix.
      Warnings are advisory.
   7. Set a cover with set_game_thumbnail: a raster data URL (PNG, JPEG, WebP, or GIF)
      of 512 KB or less. SVG is rejected, so rasterize first.
   8. Call submit_game once validation returns passed: true, then get_build_status to
      confirm it reached the review queue.
   
   Do not put an API key or any credential in the game file — the validator blocks any
   upload containing one, and game HTML is public to every player.
   
   Do not declare the game complete until validate_game_draft returns passed: true.
   
   submit_game sends the game for review. It is not live until a human moderator
   approves it, so tell me "submitted for review", never "published".
   
   Tell me every tool call you make and what it returned. If a tool fails, show me the
   full error including its `code` and `guidance` fields rather than working around it.

   To pin down what it builds, type your idea into the starter prompt card on /create and copy the version it writes. The session appears on the sessions page as soon as the agent opens it.

If something goes wrong