Skip to content

YC (S26) | Open Computer History | Continuously record your company computer work, map your workflows, help you find work worth automating, and power your agents' context

3.2 KB
Raw Blame
---
title: "Msty - privacy-first AI with screen context"
sidebarTitle: "Msty"
description: "Connect Screenpipe to Msty's Toolbox via MCP (STDIO or HTTP) so this privacy-first desktop AI can search your screen history and meeting transcripts."
icon: "brain"
---

[Msty](https://msty.studio) is a privacy-first AI platform (desktop + web) that natively supports MCP via its Toolbox feature — both STDIO and HTTP modes. Screenpipe works out of the box with Msty, giving it access to your screen history, audio transcriptions, and more.

## Setup

### STDIO mode

1. Make sure Screenpipe is running on your machine
2. Open Msty and go to **Settings** → **Toolbox**
3. Add a new tool with the following configuration:

```json
{
  "mcpServers": {
    "screenpipe": {
      "command": "npx",
      "args": ["-y", "screenpipe-mcp"]
    }
  }
}
```

4. Save and enable the tool

### HTTP mode

For a client using HTTP transport on the same computer, start the HTTP MCP server:

```bash
npx -y --package screenpipe-mcp screenpipe-mcp-http --port 3031
```

Then point Msty's Toolbox at the HTTP endpoint:

```json
{
  "mcpServers": {
    "screenpipe": {
      "url": "http://localhost:3031/mcp"
    }
  }
}
```

The default server listens on loopback only. For another machine, use the authenticated [LAN setup](/mcp-server#expose-to-your-lan). Verify [a real retrieval](/mcp-server#verify-a-real-retrieval) after connecting.

## Usage

Once configured, Msty can search your screen history and get context about what you've been working on:

```
> what was I working on this morning?

> find that documentation about async/await patterns I was reading earlier

> what error messages have I seen in my terminal today?

> search for the API response format I was looking at in the browser
```

## Example workflows

**Recall context from earlier:**
```
> I was reading a blog post about rust macros earlier today,
> search screenpipe and summarize the key points
```

**Reference meeting discussion:**
```
> search my audio transcriptions for what was discussed
> about the deployment timeline, then help me plan next steps
```

**Debug from memory:**
```
> I saw an error message flash on screen, search screenpipe
> to find it and help me fix the issue
```

**Find code examples:**
```
> search screenpipe for the python code I was looking at
> in the browser yesterday about asyncio patterns
```

## Available tools

Screenpipe provides these MCP tools to Msty:

| Tool | Description |
|------|-------------|
| `search-content` | Search screen text (accessibility-first, OCR fallback), audio transcriptions, and input |
| `activity-summary` | Lightweight overview of app usage, speakers, and recent texts for a time range |
| `search-elements` | Search structured UI elements from the accessibility tree |
| `frame-context` | Full accessibility tree, URLs, and text for a specific frame |
| `list-meetings` | List detected meetings with duration, app, and attendees |
| `export-video` | Export screen recordings as MP4 for a time range |

## Requirements

- Screenpipe running on localhost:3030 (or localhost:3031 for HTTP mode)
- Msty desktop or web app
- Node.js >= 18.0.0

Need help? [join our Discord](https://discord.gg/screenpipe).