/ Directory / Playground / Mind-Cloning-Engineering
● Community yzfly ⚡ Instant

Mind-Cloning-Engineering

by yzfly · yzfly/Mind-Cloning-Engineering

Capture a person's reasoning style — principles, taste, decision patterns — into an LLM-native Agent Skill you can converse with.

MCE (Mind Cloning Engineering) is a framework for turning a specific human's thinking style into a reusable agent skill. Instead of 'here's a persona prompt', MCE separates principles, lexicon, decision heuristics, and characteristic examples into structured layers that the model grounds in.

Why use it

Key features

Live Demo

What it looks like in practice

mind-cloning-engineering-skill.replay ▶ ready
0/0

Install

Pick your client

~/Library/Application Support/Claude/claude_desktop_config.json  · Windows: %APPDATA%\Claude\claude_desktop_config.json
{
  "mcpServers": {
    "mind-cloning-engineering-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/yzfly/Mind-Cloning-Engineering",
        "~/.claude/skills/Mind-Cloning-Engineering"
      ],
      "_inferred": true
    }
  }
}

Open Claude Desktop → Settings → Developer → Edit Config. Restart after saving.

~/.cursor/mcp.json · .cursor/mcp.json
{
  "mcpServers": {
    "mind-cloning-engineering-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/yzfly/Mind-Cloning-Engineering",
        "~/.claude/skills/Mind-Cloning-Engineering"
      ],
      "_inferred": true
    }
  }
}

Cursor uses the same mcpServers schema as Claude Desktop. Project config wins over global.

VS Code → Cline → MCP Servers → Edit
{
  "mcpServers": {
    "mind-cloning-engineering-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/yzfly/Mind-Cloning-Engineering",
        "~/.claude/skills/Mind-Cloning-Engineering"
      ],
      "_inferred": true
    }
  }
}

Click the MCP Servers icon in the Cline sidebar, then "Edit Configuration".

~/.codeium/windsurf/mcp_config.json
{
  "mcpServers": {
    "mind-cloning-engineering-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/yzfly/Mind-Cloning-Engineering",
        "~/.claude/skills/Mind-Cloning-Engineering"
      ],
      "_inferred": true
    }
  }
}

Same shape as Claude Desktop. Restart Windsurf to pick up changes.

~/.continue/config.json
{
  "mcpServers": [
    {
      "name": "mind-cloning-engineering-skill",
      "command": "git",
      "args": [
        "clone",
        "https://github.com/yzfly/Mind-Cloning-Engineering",
        "~/.claude/skills/Mind-Cloning-Engineering"
      ]
    }
  ]
}

Continue uses an array of server objects rather than a map.

~/.config/zed/settings.json
{
  "context_servers": {
    "mind-cloning-engineering-skill": {
      "command": {
        "path": "git",
        "args": [
          "clone",
          "https://github.com/yzfly/Mind-Cloning-Engineering",
          "~/.claude/skills/Mind-Cloning-Engineering"
        ]
      }
    }
  }
}

Add to context_servers. Zed hot-reloads on save.

claude mcp add mind-cloning-engineering-skill -- git clone https://github.com/yzfly/Mind-Cloning-Engineering ~/.claude/skills/Mind-Cloning-Engineering

One-liner. Verify with claude mcp list. Remove with claude mcp remove.

Use Cases

Real-world ways to use Mind-Cloning-Engineering

Capture a mentor's reasoning and have it review your work

👤 Anyone with access to a mentor's writing / talks / recorded advice ⏱ ~120 min advanced

When to use: You want that mentor's lens available even when they're not.

Prerequisites
  • Source material from the mentor — Collect essays, talk transcripts, emails, recorded conversations (with consent)
  • Skill installed — git clone https://github.com/yzfly/Mind-Cloning-Engineering ~/.claude/skills/Mind-Cloning-Engineering
Flow
  1. Extract layers
    Use Mind-Cloning-Engineering. From these essays, extract: core principles, recurring lexicon, decision heuristics, and 5 exemplar choices.✓ Copied
    → Four layered documents, not a generic persona blob
  2. Package as a skill
    Assemble into a skill folder with SKILL.md that references these layers.✓ Copied
    → Ready-to-install skill directory
  3. Consult the clone
    I'm considering [decision]. Review through the lens of this clone — what would the heuristics and principles suggest?✓ Copied
    → Specific references to principles, not generic 'consider...'

Outcome: A reusable advisor grounded in real material, not a vibe.

Pitfalls
  • Treating the clone as the real mentor — It's an approximation. Use for framing, not final calls. Respect the real person's actual current views.
Combine with: filesystem

Externalize your own thinking style as a checklist skill

👤 Senior practitioners wanting to codify their own taste ⏱ ~60 min intermediate

When to use: You keep making the same kinds of good decisions and want them explicit.

Flow
  1. Mine your own artifacts
    Use Mind-Cloning-Engineering on my past design docs and PR comments in docs/. Extract my recurring principles and heuristics.✓ Copied
    → A structured doc of your own patterns
  2. Use as a self-review tool
    Review this draft using the clone of my own principles. What would I flag if a teammate submitted this?✓ Copied
    → Concrete flags rooted in your own rules

Outcome: You catch your own blind spots from a second angle.

Pitfalls
  • Clone becomes an echo chamber — Periodically refresh with new material; your thinking evolves

Combinations

Pair with other MCPs for X10 leverage

mind-cloning-engineering-skill + filesystem

Organize multiple clones in a library

Maintain clones/ with one folder per cloned mind.✓ Copied

Tools

What this MCP exposes

ToolInputsWhen to callCost
extract_principles source documents Starting a clone 0
extract_lexicon source documents Voice capture 0
extract_heuristics decisions + outcomes Capturing decision patterns 0
assemble_skill all layers Packaging 0

Cost & Limits

What this costs to run

API quota
None
Tokens per call
High during extraction (reads lots of source), low during consultation
Monetary
Free
Tip
Extract once carefully, consult cheaply for months.

Security

Permissions, secrets, blast radius

Credential storage: None
Data egress: Source material goes through Claude API during extraction

Troubleshooting

Common errors and fixes

Clone feels generic

Source material too thin or too formal; add more informal artifacts (emails, chat, off-the-cuff)

Clone contradicts the real person

Flag and update; principles should be refreshed periodically

Alternatives

Mind-Cloning-Engineering vs others

AlternativeWhen to use it insteadTradeoff
System prompt personaYou only need a quick vibe matchNo structural grounding, drifts fast
Fine-tuning a model on the personYou have lots of data and deep pocketsExpensive, opaque, harder to edit

More

Resources

📖 Read the official README on GitHub

🐙 Browse open issues

🔍 Browse all 400+ MCP servers and Skills