Skill

yutori-browse

yutori-browse — a Claude Code skill from yutori-ai/yutori-mcp

Features

✓**Scouting** — Monitor the web continuously for anything you care about at a desired frequency
✓**Research** — Run one-time deep web research tasks
✓**Browsing** — Automate websites with an AI navigator
✓[`/yutori-scout`](skills/01-scout/SKILL.md) — Set up continuous web monitoring
✓[`/yutori-research`](skills/02-research/SKILL.md) — Deep web research (async, 5–10 min)
✓[`/yutori-browse`](skills/03-browse/SKILL.md) — Browser automation
✓[`/yutori-competitor-watch`](skills/04-competitor-watch/SKILL.md) — Competitor monitoring template
✓[`/yutori-api-monitor`](skills/05-api-monitor/SKILL.md) — API/changelog monitoring template

Installation Guide

<details>
<summary>Requirements</summary>

If you don't already have uv installed, install it (it includes uvx):

bash

curl -LsSf https://astral.sh/uv/install.sh | sh

Or with Homebrew:

bash

brew install uv

Python 3.10 or higher is required (uv manages this automatically for most installs).

For the quickstart below, Node.js is also required (for npx).
</details>

Quick install (recommended)

![MCP server installation](assets/mcp-server-install.gif)

Run in terminal:

bash

    uvx yutori-mcp login

This will open Yutori Platform in your browser and save your API key locally.

<details>
<summary>Or, manually add your API key</summary>

Go to (https://platform.yutori.com) and add your key to the config file:

bash

    mkdir -p ~/.yutori
    cat > ~/.yutori/config.json << 'EOF'
    {"api_key": "yt-your-api-key"}
    EOF

</details>

Install MCP using add-mcp (requires Node.js):

text

   npx add-mcp "uvx yutori-mcp"

Pick the clients you want to configure.

Install workflow skills using skills.sh (requires Node.js):

text

   npx skills add yutori-ai/yutori-mcp -g

Adds slash-command shortcuts like /yutori-scout, /yutori-research, and more.

-g installs them at user scope. Omit -g if you want a project-local install instead.

<details>
<summary>To list or remove skills later:</summary>

bash

   npx skills ls -g
   npx skills remove -g yutori-login

</details>

Restart the tool you are using.

Manual per-client setup

<details>
<summary>Claude Code</summary>

Plugin (Recommended) - Includes MCP tools + workflow skills

Type these commands in Claude Code's input (not in a terminal):

text

   /plugin marketplace add yutori-ai/yutori-mcp
   /plugin install yutori@yutori-plugins

This installs both the MCP tools and workflow skills:

| Skill | Description |
|-------|-------------|
| /yutori-scout | Set up continuous web monitoring with comprehensive queries |
| /yutori-research | Deep web research workflow (async, 5-10 min) |
| /yutori-browse | Browser automation tasks |
| /yutori-competitor-watch | Quick competitor monitoring template |
| /yutori-api-monitor | API/changelog monitoring template |

> Already have the MCP server installed? Remove it first to avoid duplicate configurations:
>

bash

   > claude mcp remove yutori -s user   # if installed at user scope
   > claude mcp remove yutori -s local  # if installed at local/project scope
   >

To uninstall the plugin later:

text

   /plugin uninstall yutori@yutori-plugins -s user

MCP Only (if you prefer not to use the plugin)

bash

   claude mcp add --scope user yutori -- uvx yutori-mcp

The server reads your API key from ~/.yutori/config.json (set up via uvx yutori-mcp login).
</details>

<details>
<summary>Claude Desktop</summary>

Add to your claude_desktop_config.json:

json

{
  "mcpServers": {
    "yutori": {
      "command": "uvx",
      "args": ["yutori-mcp"]
    }
  }
}

The server reads your API key from ~/.yutori/config.json.

For setup details, see the Claude Desktop MCP install guide.
</details>

<details>
<summary>Cursor</summary>

Click the button to install:

Or install manually:

Go to Cursor Settings → MCP → Add new MCP Server, then add:

json

{
  "mcpServers": {
    "yutori": {
      "command": "uvx",
      "args": ["yutori-mcp"]
    }
  }
}

The server reads your API key from ~/.yutori/config.json.

See the Cursor MCP guide for setup details.
</details>

Click the button to install:

Or install manually:

bash

code --add-mcp '{"name":"yutori","command":"uvx","args":["yutori-mcp"]}'

The server reads your API key from ~/.yutori/config.json.
</details>

<details>
<summary>ChatGPT</summary>

Open ChatGPT Desktop and go to Settings -> Connectors -> MCP Servers -> Add server.

json

{
  "mcpServers": {
    "yutori": {
      "command": "uvx",
      "args": ["yutori-mcp"]
    }
  }
}

The server reads your API key from ~/.yutori/config.json.

For setup details, see the OpenAI MCP guide.
</details>

<details>
<summary>Codex</summary>

MCP Server:

bash

   codex mcp add yutori -- uvx yutori-mcp

Or add to ~/.codex/config.toml:

toml

   [mcp_servers.yutori]
   command = "uvx"
   args = ["yutori-mcp"]

The server reads your API key from ~/.yutori/config.json.

Skills (optional, for workflow guidance):

Install skills using $skill-installer inside Codex:

text

   $skill-installer install https://github.com/yutori-ai/yutori-mcp/tree/main/.agents/skills/yutori-scout
   $skill-installer install https://github.com/yutori-ai/yutori-mcp/tree/main/.agents/skills/yutori-research
   $skill-installer install https://github.com/yutori-ai/yutori-mcp/tree/main/.agents/skills/yutori-browse
   $skill-installer install https://github.com/yutori-ai/yutori-mcp/tree/main/.agents/skills/yutori-competitor-watch
   $skill-installer install https://github.com/yutori-ai/yutori-mcp/tree/main/.agents/skills/yutori-api-monitor

Or manually copy skills to your user directory (use -L so symlinks are dereferenced and real files are copied):

bash

   git clone https://github.com/yutori-ai/yutori-mcp /tmp/yutori-mcp
   cp -rL /tmp/yutori-mcp/.agents/skills/* ~/.agents/skills/

To uninstall manually copied skills, delete the matching directories from ~/.agents/skills/. When updating this way, remove old Yutori skill directories first, since cp -rL will not delete renamed or removed skills.

Restart Codex after installing skills.

| Skill | Command | Description |
|-------|---------|-------------|
| Scout | $yutori-scout | Set up continuous web monitoring |
| Research | $yutori-research | Deep web research (async, 5-10 min) |
| Browse | $yutori-browse | Browser automation with AI navigator |
| Competitor Watch | $yutori-competitor-watch | Quick competitor monitoring template |
| API Monitor | $yutori-api-monitor | API/changelog monitoring template |

See the Codex Skills docs for more on skills.
</details>

<details>
<summary>OpenClaw</summary>

Follow the Quickstart above:

Install skills and MCP for OpenClaw (and optionally other tools) via skills.sh:

bash

   npx skills add yutori-ai/yutori-mcp

When prompted, choose which Yutori skills to install and select OpenClaw as the tool.

</details>

<details>
<summary>Gemini CLI</summary>

Add to ~/.gemini/settings.json. If you already have mcp or mcpServers, merge these keys into your existing config:

json

{
  "mcp": {
    "allowed": ["yutori"]
  },
  "mcpServers": {
    "yutori": {
      "command": "uvx",
      "args": ["yutori-mcp"]
    }
  }
}

The server reads your API key from ~/.yutori/config.json.

Add "yutori" to mcp.allowed if you already list other MCPs there. For more details, see the Gemini CLI MCP settings guide.
</details>

Install the package to run the MCP server (e.g. for custom or self-hosted setups):

bash

pip install yutori-mcp

</details>

yutori-browse

Features

Installation Guide

Quick install (recommended)

Manual per-client setup

Reviews