• LABELED
  • SCORED
  • UNCERTAIN
  • FILTERED

How to classify text and images with your AI

Connect Nyckel to your AI client once. From then on, your agent uses your trained models to label incoming data.

Ask AI about this page

Short answer

How can I use my AI to classify text and images?

Your AI calls your trained Nyckel models to get instant classifications and confidence scores for new data. You'll see how the agent handles these predictions below.

Where your data lands.

The results of your predictions.

The AI sorts your data based on the labels you've already trained in Nyckel.

LABELED

Categorized data

The AI identifies the correct label for a text string or image. You get a clean classification to use in your next step.

SCORED

Confidence levels

Every prediction includes a score. You can see exactly how sure the model is about its choice.

UNCERTAIN

Human review needed

If the confidence score is too low, the AI flags the item. You decide if a person needs to step in and check it.

FILTERED

Cleaned datasets

The AI separates relevant data from noise. This keeps your downstream processes focused only on what matters.

The workflow

What your AI does when data arrives.

Your agent acts as the bridge between your raw data and your trained Nyckel models.

  1. Find available models

    The AI checks which classification models you have ready to use.

    list_ml_functions

  2. Check available labels

    The agent looks up the specific categories your model is trained to recognize.

    list_ml_labels

  3. Run predictions

    The AI sends your text or image to the model to get a real-time result.

    invoke_ml_function

  4. Process scores

    The AI reads the returned labels and confidence scores to determine the next action.

    invoke_ml_function

Try it

Copy these to start.

Use these prompts to begin working with your Nyckel models.

Starting points

These are opening lines. Swap out the specific data sources like a Slack channel or a Google Sheet with your own.

Nyckel Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_XHJXGFWpxs8iH5SOEbAtipKgPVXhxGte0LRcKzdL/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Nyckel capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "nyckel-ml-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_XHJXGFWpxs8iH5SOEbAtipKgPVXhxGte0LRcKzdL/mcp"
    }
  }
}
Copy into chat04

  • Use my Nyckel models to classify the latest messages in this Slack channel.

  • Check the images in my Google Drive folder and label them using my trained model.

  • Look at the new rows in this spreadsheet and run a prediction on the text column.

  • List my available Nyckel functions so I know which models I can use.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Nyckel once, then ask.

Click the link to connect your account. Your credentials stay encrypted, and you can immediately start asking your AI to run predictions.

Connect Nyckel to your AI

FAQ

How this task behaves.

  • 01

    Can the AI change my Nyckel model settings?

    No. The AI can only use your existing models to make predictions; it cannot edit or retrain them.

  • 02

    Does the AI delete any data in Nyckel?

    No. The AI only reads model information and runs predictions. It has no capability to delete your data or models.

  • 03

    Can I use this for both text and images?

    Yes, as long as you have trained models in Nyckel that are designed for those specific data types.

  • 04

    How do I know if a prediction is accurate?

    The AI receives a confidence score with every prediction, which you can use to decide if the result is reliable.

  • 05

    Can the AI create new labels for me?

    No. The AI can only use the labels that already exist in your trained Nyckel models.

  • More questions about Nyckel? See everything the Nyckel Connector can do

Connect Nyckel to Claude, Cursor, ChatGPT & more