# emulo

Emulo parses your AI coding session logs and builds a you.md profile with your actual working rules — each one tied to a date and your own message. Your agent can read it and work your way from the start.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 306 repo stars
- Author: ohad6k
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/ohad6k/emulo
- Updated: 2026-10-12
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/emulo

## What it does

Emulo reads raw logs from Claude Code, Codex, Copilot CLI, OpenCode and similar platforms to find the patterns you live by: what you reject on sight, what "done" means, how you debug, how you write UI. Each rule in the profile carries a timestamp and a quote from your chat — it is not a template, it is what actually worked.

The profile splits into layers: work, design, writing, and video. A host loads only the layer that fits the task. Emulo does not claim the profile will make your agent work better (that is hard to measure), but in September 2026 tests it changed model output on real tasks. Support varies by platform: in Codex the profile reaches the model through `AGENTS.md`, in Claude Code you type `/you` to load it by hand, in Cursor and Gemini the file lands where they look for it.

Installing the server and actually loading the profile are separate steps. Python with no external dependencies is all you need. Run the mine command, select what to include from your logs, and the profile builds on your machine. If you point mining at a remote model, selected evidence goes there; a local model keeps everything local.

## Requirements
- Needs: Python and uv (commands run via uvx)

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add emulo -- uvx emulo
```
uvx runs the command named after the package; if the README names another one, use uvx --from emulo <command>

### Codex

**Platform CLI**
```bash
codex mcp add emulo -- uvx emulo
```
uvx runs the command named after the package; if the README names another one, use uvx --from emulo <command>

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "emulo": {
            "command": "uvx",
            "args": [
                "emulo"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from emulo <command>

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "emulo": {
            "type": "stdio",
            "command": "uvx",
            "args": [
                "emulo"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from emulo <command>

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add emulo uvx emulo
```
uvx runs the command named after the package; if the README names another one, use uvx --from emulo <command>

## When to use it
- extract coding rules and patterns from your chat history with agents
- build a dated profile with real examples instead of generic templates
- give your agent context about how you work and what you reject
- split rules by work type — code, design, video, writing
- load the profile into your IDE so the model can read it

## Example prompts
- Run `/emulo:mine` in Claude Code, pick your session logs, and choose which parts to add to the profile.
- After `/plugin install emulo@emulo` in Codex, call `emulo:mine`, point it at your log folder, and let it build the profile.
- Run `emulo:mine` with a local model to keep the whole process on your machine, then copy `you.md` into your project.

## Similar
- [Alan McIntyre](https://affpapa.org/en/nevibecode/mcp/alan-mcintyre.md) — MCP server · 377 installs
- [conversation-archaeologist](https://affpapa.org/en/nevibecode/skills/conversation-archaeologist.md) — skill · 371 installs
- [setup](https://affpapa.org/en/nevibecode/skills/proficientlyjobs-setup.md) — skill · 118 installs
- [theabbie](https://affpapa.org/en/nevibecode/mcp/theabbie.md) — MCP server · ★ 65 repo stars
- [mcp-me](https://affpapa.org/en/nevibecode/mcp/mcp-me.md) — MCP server · ★ 21 repo stars
- [Entry](https://affpapa.org/en/nevibecode/mcp/entry.md) — MCP server

From the same set: [ohad6k/emulo](https://affpapa.org/en/nevibecode/by/ohad6k/emulo.md) — 7 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
