• ESSENTIAL
  • DEBATABLE
  • BLOAT
  • NOISE

How to prune product features with your AI

Connect your AI to your product specs and force it to defend every single line of code. If a feature can't justify its existence through subtraction, it goes.

Ask AI about this page

Short answer

How can I use AI to cut feature bloat?

You use your AI to run a ruthless audit where it must prove a feature's worth by simulating its removal. This process exposes unnecessary complexity and highlights what actually matters to your users.

The results of the audit.

Where your feature list lands.

The AI categorizes every feature based on its necessity and impact.

ESSENTIAL

Core product pillars

The AI identifies features that are fundamental to the user's primary goal. These stay because they are the reason the product exists.

DEBATABLE

Requires human judgment

These are features that provide value but add complexity. You'll need to step in here to decide if the trade-off is worth it.

BLOAT

Candidates for removal

The AI flags features that fail to justify their presence through subtraction. These are the first targets for your next cleanup.

NOISE

Pure distraction

The AI isolates features that actively clutter the interface or confuse the user journey. These are identified as high-priority cuts.

The workflow

What your AI does when the feature list arrives.

The AI acts as a ruthless editor, stripping away anything that doesn't serve the core vision.

  1. Analyze the spec

    The AI reads your current feature list or product documentation to understand the baseline.

    validate_steve_jobs_vision

  2. Apply subtraction logic

    It tests the product's utility by simulating the removal of specific components.

    validate_steve_jobs_vision

  3. Identify bloat

    The AI flags features that add weight without adding proportional value to the core experience.

    validate_steve_jobs_vision

  4. Generate a lean roadmap

    It produces a list of what to keep and what to cut to reach a more focused product state.

    validate_steve_jobs_vision

Try it

Copy these to start.

Use these prompts to kick off the pruning process.

Starting points

Swap the bracketed text for your actual product docs or feature lists.

Steve Jobs Vision Prover Connector

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

Connector linkhttps://edge.vinkius.com/vk_preview_utuErD2gnpH4HqP1nZAn2SOTR0g5vHhqwe0DBzRx/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 — Steve Jobs Vision Prover capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "steve-jobs-vision-prover-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_utuErD2gnpH4HqP1nZAn2SOTR0g5vHhqwe0DBzRx/mcp"
    }
  }
}
Copy into chat04

  • Analyze this feature list from [Product Name] and tell me which parts are just bloat: [Paste List]

  • Look at my product requirements in [Document Name] and identify features that don't serve the core user goal.

  • If I removed [Feature Name] from my current roadmap, would the core value of the product decrease? Justify your answer.

  • Review the feature set in [Repo/Doc] and find everything that makes the user experience more confusing.

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

Start here

Connect Steve Jobs Vision Prover once, then ask.

Link your preferred AI client to the MCP. Your credentials stay encrypted, and you can start pruning immediately.

Connect Steve Jobs Vision Prover to your AI

FAQ

How this task behaves.

  • 01

    Can the AI actually delete code from my repository?

    No. The AI only analyzes your data and provides recommendations. You must manually perform any deletions.

  • 02

    Does the AI decide which features to keep?

    No. The AI provides the logic and the justification, but the final decision on what stays or goes is always yours.

  • 03

    What kind of files can I use for this?

    You can use any text-based documentation, feature lists, or product requirement documents you provide to your AI client.

  • 04

    How does it 'justify through subtraction'?

    The AI evaluates the product's primary function and determines if a feature is essential to that function or just an extra layer of complexity.

  • 05

    Will it work with my existing product docs?

    Yes, as long as you can provide that text or file to your AI client, the MCP allows the AI to process it.

  • More questions about Steve Jobs Vision Prover? The Connector page answers them. See everything the Steve Jobs Vision Prover Connector can do

Connect Steve Jobs Vision Prover to Claude, Cursor, ChatGPT & more