# ainative-browser-mcp [Health: Active]

**Category:** 📂 Browser Automation  
**Repository:** https://github.com/AINative-Studio/ainative-browser-mcp  
**GitHub Stars:** 0  
**npm Downloads (last month):** 1562  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/ainative-browser-mcp

## Description
AI-powered browser automation MCP — 6 tools for web interaction, extraction, and validation.

## Tools
Capabilities this server exposes over MCP:

- **browser_act** — Perform an action on a web page using AI-powered browser automation. Use when you need to click, type, navigate, or otherwise interact with a page. (50 credits)
- **browser_extract** — Extract structured data from a web page. Use when you need to scrape, parse, or retrieve specific information from a URL. Returns data in structured JSON format. (75 credits)
- **browser_validate** — Validate content or state on a web page. Use when you need to assert that specific content exists, a form was submitted, or a page reached an expected state. (25 credits)
- **browser_task** — Run a multi-step browser automation task. Use when you need to perform a complex, multi-step workflow on a web page (e.g. login and fill out a form). More capable than browser_act for complex flows. (200 credits)
- **browser_extract_to_table** — Extract structured data from a web page and store it directly in a ZeroDB NoSQL table. Use when you need to scrape data and persist it for later querying or analysis. (100 credits)
- **browser_enrich_memory** — Extract content from a web page and store it in agent memory (ZeroDB). Use when you want an agent to "remember" information from a URL for future recall. (100 credits)
- **browser_batch_extract** — Extract structured data from up to 10 URLs in a single call and store rows in a ZeroDB table. Use when you need to scrape multiple pages at once. (75 credits per URL)
- **browser_enrich_memory_async** — Queue an async browser memory enrichment task. Returns a task_id immediately — the browser visits the URL and stores content in ZeroMemory in the background. (100 credits)

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "ainative-browser-mcp": {
    "command": "npx",
    "args": ["-y","@ainative/browser-mcp"],
    "env": {
      "AINATIVE_API_KEY": "",
      "AINATIVE_USERNAME": "",
      "AINATIVE_PASSWORD": ""
    }
  }
}
```

**Requires environment variables:** `AINATIVE_API_KEY`, `AINATIVE_USERNAME`, `AINATIVE_PASSWORD` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What ainative-browser-mcp does

The ainative-browser-mcp MCP server exposes browser automation functions to AI agents through six MCP tools. Agents can navigate to a URL, click elements, enter text, retrieve structured information, and check whether content or page state meets a requested condition.

The tool set also covers longer workflows and storage. `browser_task` runs a multi-step automation request, while `browser_extract_to_table` writes extracted results to a ZeroDB table. `browser_enrich_memory` extracts page content and stores it in agent memory, using a supplied memory type and project ID.

## How it works

An MCP client starts the package with `npx @ainative/browser-mcp`. Tool calls include a target URL and an instruction or extraction goal, depending on the operation. For example, an extraction request can ask for product names and prices, while an action request can direct the agent to click a sign-in button and set a maximum number of steps.

The server chooses its backend automatically. When ZeroLocal is running at `localhost:8000`, it uses that local service. If ZeroLocal is unavailable, requests use AINative Cloud instead. ZeroLocal can be installed with `pip install zerodb-local` and started with `zerodb serve`.

## Setup and configuration

Install the package directly for an on-demand launch:

```bash
npx @ainative/browser-mcp
```

A global npm installation is also supported:

```bash
npm install -g @ainative/browser-mcp
```

Authentication can use an API key or a username and password. Set `AINATIVE_API_KEY` for the recommended method, or provide `AINATIVE_USERNAME` and `AINATIVE_PASSWORD`. The API key is obtained from the AINative Studio dashboard.

For Claude Code, add the package as an MCP server with `npx` as the command and `@ainative/browser-mcp` as its argument. Include the chosen authentication variables in the server environment. No extra configuration is needed for ZeroLocal auto-detection.

## Tools and capabilities

The ainative-browser-mcp MCP server provides:

- `browser_act`: Performs page actions such as clicking, typing, and navigation.
- `browser_extract`: Returns structured data based on an extraction goal.
- `browser_validate`: Checks page content or state.
- `browser_task`: Runs a multi-step browser automation task.
- `browser_extract_to_table`: Extracts information and stores it in a ZeroDB table.
- `browser_enrich_memory`: Extracts content and saves it to agent memory.

The README assigns credit costs to each tool: 50 for actions, 75 for extraction, 25 for validation, 200 for multi-step tasks, and 100 each for table extraction and memory enrichment. These costs may matter when choosing between individual operations and larger tasks.

## Limitations and notes

The package requires authentication unless a supported local setup supplies the required backend behavior. The documented local option is ZeroLocal at `localhost:8000`; otherwise the server uses AINative Cloud. Storage-oriented tools depend on the relevant ZeroDB table or agent-memory context, including a project ID for the documented memory example.

Those details should be confirmed in the tool interface or project documentation before building a workflow around them.

_Full upstream README: https://allmcps.com/mcp/ainative-browser-mcp/readme_