# Persome

Persome keeps a local model of your working context on Mac and exposes it to trusted AI clients. Use it when several agents need the same personal context.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 1328
- Author: Intuition-Lab
- License: Apache-2.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/Intuition-Lab/personal-model
- Updated: 2026-10-16
- HTML: https://affpapa.org/en/nevibecode/mcp/persome

## What it does

Persome is a local MCP server for long-term personal memory. It learns from focused activity on a Mac after system access is approved, then gives connected agents context about goals, priorities, and working patterns. The model stays on the Mac and can be inspected, corrected, exported, or deleted.

Its context model moves from sourced observations through relationships and patterns to a current Root. Evidence stays attached to important claims, so later activity can revise an earlier inference. The same model can serve Claude Code, Codex, Cursor, and other trusted MCP clients.

The server runs locally with `uvx personal-model`. Persome depends on macOS permission to learn from selected activity, so that permission is required before this part can work.

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

## Install

### Claude Code (native platform)

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

### Codex

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

### Cursor

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

### GitHub Copilot

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

### Gemini CLI

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

## When to use it
- share one personal context across coding agents
- carry goals and priorities between AI clients
- trace claims back to supporting observations
- correct or remove stored personal context
- connect local memory through MCP

## Example prompts
- Use my current priorities when you plan the next coding task.
- Compare this task with my usual working patterns and suggest a next step.
- Show which observations support that conclusion.

## Similar
- [mem0](https://affpapa.org/en/nevibecode/skills/mem0.md) — skill · 2300 installs
- [mem0-cli](https://affpapa.org/en/nevibecode/skills/mem0-cli.md) — skill · 773 installs
- [Agentic UX Design - Relationship-Centric Interfaces](https://affpapa.org/en/nevibecode/skills/agentic-ux-design-relationship-centric-interfaces.md) — skill · 3300 installs
- [speak-tts](https://affpapa.org/en/nevibecode/skills/speak-tts.md) — skill · 735 installs
- [apple-intelligence](https://affpapa.org/en/nevibecode/skills/apple-intelligence.md) — skill · 516 installs

---
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