Skip to main content
Claude Code · Cursor · VS Code · Claude.ai

Ask Claude — get images, video, and audio back

Connect the core.today MCP server and Claude searches models, estimates cost, generates, and fetches results — no code required, just conversation.

Connect core.today to Claude

Follow the steps below and Claude can start generating images, video, and audio.

Loading…
This command contains your API key. Be careful with screen sharing, screenshots, and commits — corporate networks may also log it in proxies. We recommend issuing a dedicated key for MCP. The "Add" buttons pass the key-bearing config through the cursor.com (or vscode.dev) redirector — if the app is installed, the "Open directly in the app" button is the path that keeps the key local.

Choose a key. Until then the command shows a placeholder key.

Add the Core.Today MCP server to Cursor settings in one click.

Set up manually

{
  "mcpServers": {
    "coretoday": {
      "url": "https://api.core.today/v1/mcp",
      "headers": {
        "X-API-Key": "cdt_YOUR_KEY"
      }
    }
  }
}

Verify the connection

Ask Claude

What's my core.today balance?

get_balance is free, read-only, and instant, so this one line proves the whole auth path works end to end.

Your first prompt

Make an image with core.today. First use search_models to find a cheap model, then estimate_credits to tell me the cost, then go ahead.

Without this, Claude will call create_prediction on the first model it notices.

Money and safety

  • create_prediction draws on your team's shared credits — any teammate's Claude can spend the team balance.
  • An API key's "allowed models / allowed IPs" settings are not enforced on the MCP path. The only guardrail that's actually enforced is the rate limit.
  • Failed calls are refunded automatically, but successful ones are not. A video model can cost 7,400+ credits per call, and the daily free allowance is 50.

9 tools, Claude calls them for you

Claude calls these tools on its own — nothing to memorize.

Find models

search_models

Search for models that match what you need.

No credits

Check model spec

get_model_schema

Look up a model's input parameter schema.

No credits

Preview cost

estimate_credits

Calculate the expected credit cost before running.

No credits

Generate image/video

create_prediction

Generate images, video, audio, and other AI output.

Uses credits

Check result

get_prediction

Check the status and result of a generation job.

No credits

Cancel generation

cancel_prediction

Cancel a generation job in progress.

No credits

Upload a file

create_upload_url

Get a URL for uploading a file.

No credits · storage quota

Import file from URL

import_file_from_url

Fetch and register a file from an external URL.

No credits · storage quota

Check balance

get_balance

Check your team's current credit balance.

No credits

This is what a real conversation looks like

Below is the exact tool-call sequence Claude runs.

Generate an image

Make an image of a dog playing on a beach at sunset
search_models

Called search_models for image generation — found 12 models including nano-banana-pro

estimate_credits

Called estimate_credits — about 120 credits

create_prediction

Called create_prediction to start generating

Done! 120 credits were deducted.

Edit an image from a URL

Change the background to the ocean in this image: https://example.com/photo.jpg
import_file_from_url

Called import_file_from_url to fetch the image — storage quota only, no credits used

create_prediction

Called create_prediction with an edit model (nano-banana-pro/edit)

Changed the background to the ocean.

Frequently asked questions

Connect to Claude now

Issue an API key and follow the install steps above — your first result can be ready in minutes.

Get an API keyRead the reference docs