About noncelogic
noncelogic is an MCP server for Rove’s hosted Playwright API, designed for AI agents performing browser tasks. It returns accessibility trees alongside browser controls, giving agents structured page data rather than relying only on screenshots.
Key features
- 01Navigate to URLs and manage sessions
- 02Click and fill page elements
- 03Extract data with JSON schemas
- 04Retrieve accessibility trees
- 05Capture screenshots
Use cases
- 01Automating browser navigation for an AI agent
- 02Extracting structured data from web pages
- 03Inspecting page content through accessibility snapshots
MCP support
Exposes Rove browser automation capabilities to MCP clients, including navigation, interaction, extraction, screenshots, and accessibility-tree retrieval.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y @roveapi/mcpROVE_API_KEYSecretYour Rove API key (get one at roveapi.com)Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
15Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| url | string | Yes | Full URL to navigate to (e.g. "https://example.com") |
| stealth | boolean | No | Enable stealth mode: sets a realistic user agent and hides the webdriver flag to reduce bot detection on public sites. Default: false. |
| session_id | string | No | Reuse an existing browser session. If omitted, a new session is created automatically. |
| timeout_ms | number | No | Navigation timeout in milliseconds. Default: 30000. |
| wait_until | enum | No | Page lifecycle event to wait for. Default: "networkidle".load | domcontentloaded | networkidle |
| action_delay_ms | object | No | Add random delay between browser actions for polite scraping. Specify min and max milliseconds, e.g. {"min": 500, "max": 1500}. |
| action_delay_ms.max | number | No | Maximum delay in milliseconds |
| action_delay_ms.min | number | No | Minimum delay in milliseconds |
Details
From the project README@roveapi/mcp
MCP server for Rove — a hosted Playwright API for AI agents.
Returns accessibility trees instead of screenshots, reducing LLM token consumption by ~77% (26K tokens vs 114K for a typical page).
Quick Start
npx -y @roveapi/mcp
Set your API key via environment variable:
ROVE_API_KEY=rvp_live_... npx -y @roveapi/mcp
Install in Claude Code
claude mcp add --scope user rove -e ROVE_API_KEY=rvp_live_YOUR_KEY -- npx -y @roveapi/mcp
Install in Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"rove": {
"command": "npx",
"args": ["-y", "@roveapi/mcp"],
"env": {
"ROVE_API_KEY": "rvp_live_YOUR_KEY"
}
}
}
}
Install in Cursor
Add to .cursor/mcp.json:
{
"mcpServers": {
"rove": {
"command": "npx",
"args": ["-y", "@roveapi/mcp"],
"env": {
"ROVE_API_KEY": "rvp_live_YOUR_KEY"
}
}
}
}
Tools
| Tool | Description |
|---|---|
navigate | Navigate to a URL. Auto-creates a session. Supports stealth mode and action jitter. |
interact | Click or fill actions in the active session. |
extract_schema | Extract structured data from a URL using a JSON schema. |
screenshot | Take a screenshot (in-session or standalone). |
get_a11y_tree | Get the accessibility tree snapshot — the core differentiator. |
close_session | Close a browser session and release resources. |
Why Accessibility Trees?
Traditional browser automation returns screenshots that cost ~114K tokens per page. Rove returns structured accessibility trees at ~26K tokens — 77% fewer tokens, faster responses, and better structured data for LLMs to reason over.
Configuration
| Environment Variable | Default | Description |
|---|---|---|
ROVE_API_KEY | rvp_live_demo | Your Rove API key |
ROVE_API_BASE_URL | https://api.roveapi.com | API endpoint |
Get an API Key
Sign up at roveapi.com — 100 free credits on signup, no card required.
License
MIT
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.