# ShadowBot MCP for AI Agents AI Agent Connect

> ShadowBot MCP lets you manage your RPA robots and automate browser tasks directly from your AI agent. It handles everything from starting new jobs and checking robot status to pulling execution logs and managing organizational departments. This connection gives you a way to oversee your digital workforce without leaving your main workspace.

## Overview
- **Category:** productivity
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_ufnlZ9qwI42evVPv5I9DUys2Zu6x1WXTg98eorvf/ai-agent-connect
- **Tags:** shadowbot, rpa, robotic-process-automation, workflow-orchestration, browser-automation, digital-workforce

## Description

Connecting your AI agent to ShadowBot changes how you handle repetitive tasks. Instead of manually logging into a dashboard to check if a browser automation finished or to start a new batch of data entry, you can just tell your agent what to do. You'll get status updates, execution logs, and robot health reports directly in your chat. This Connector is part of the Vinkius catalog, which makes it easy to plug into your existing workflow without setting up complex custom scripts. It's about moving the control of your digital workforce into the same place you're already doing your thinking. You can see which robots are online, manage department members, and trigger specific workflows with natural language commands. It turns a fragmented RPA setup into a cohesive set of actions your agent can actually execute. You get a centralized way to oversee your entire automation pipeline, from high-level project structures down to the individual logs of a single task. This means fewer tabs, less manual clicking, and a much faster way to keep your digital workforce running on schedule. You can oversee your entire automation pipeline from a single chat interface, allowing you to scale your operations without adding more manual oversight.

## Tools

### list_task_logs
Retrieves the history of logs for a specific task to troubleshoot errors. See exactly where a bot failed.

### stop_task
Immediately halts a running ShadowBot RPA task. Use this to kill a stuck process or stop a job early.

### get_robot_details
Pulls specific data about a single robot's configuration and status. Useful for checking individual bot health.

### get_task_details
Shows the current progress and results of a specific ShadowBot task. Use this to see if a job finished or hit a snag.

### list_apps
Shows every RPA application currently registered in your account. Good for a quick inventory of your available automations.

### list_departments
Pulls the full list of organizational departments. Helps in managing large-scale enterprise RPA structures.

### list_department_members
Shows who belongs to a specific department in your RPA structure. Use this to audit team access.

### list_online_robots
Identifies which robots are currently active and ready to work. Helps you find available capacity instantly.

### list_robots
Lists every robot associated with your account for easy identification. Provides a complete view of your fleet.

### start_task
Triggers a remote execution of a ShadowBot RPA application. Use this to kick off new jobs with one command.

## Prompt Examples

**Prompt:** 
```
Are any robots online right now?
```

**Response:** 
```
Checking your ShadowBot fleet... I found 4 robots currently online. Here they are:

| Robot ID | Status |
| :--- | :--- |
| WebScraper-01 | Online |
| DataEntry-02 | Online |
| InvoiceProc-03 | Online |
| MailBot-04 | Online |

Would you like me to start a task on one of them?
```

**Prompt:** 
```
Start the invoice processing job on robot 777.
```

**Response:** 
```
Starting the 'invoice_processing' workflow on robot 777... Success. Job ID `job_456` is now running. I'll keep an eye on it for you.
```

**Prompt:** 
```
Why did the last web scraping task fail?
```

**Response:** 
```
I've pulled the logs for your last web scraping task. It looks like it hit a timeout on the login page. 

**Error Summary:**
* **Error Type:** TimeoutException
* **Timestamp:** 2026-07-10 14:45:01
* **Message:** Connection to login.portal.com timed out after 30 seconds.

Would you like me to try starting it again?
```

## Capabilities

### Trigger RPA tasks remotely
Start a browser or desktop automation job with a simple command.

### Check robot status
See which robots are currently online and ready to work.

### Fetch execution logs
Retrieve detailed history and error reports for specific tasks.

### Manage department members
View and organize team access within your RPA structure.

### Monitor robot health
Get real-time updates on your digital workforce performance.

### List account applications
View all registered RPA applications in your account.

## Use Cases

### Automated Web Scraping
An analyst asks the agent to start a web scraping task. The agent uses start_task and notifies the user when the data is ready.

### Fleet Health Check
An ops manager asks 'How many robots are online?' The agent calls list_online_robots and provides a count of active bots.

### Error Debugging
A dev asks why a job failed. The agent calls list_task_logs for that Job ID and summarizes the specific error for them.

### Team Access Audit
A manager asks to see who is in the 'Finance' department. The agent uses list_department_members to provide the names.

## Benefits

- Remote Triggering: Use start_task to kick off browser automations without leaving your current workspace.
- Real-time Monitoring: Check list_online_robots to see your digital workforce availability at a glance.
- Easy Troubleshooting: Use list_task_logs to see exactly where an automation failed without digging through a separate console.
- Org Management: Use list_departments and list_department_members to manage your team's RPA access programmatically.
- Status Tracking: Get instant results via get_task_details to know if a job finished or hit a snag.

## How It Works

The bottom line is you get a direct command line for your RPA bots inside your AI client.

1. Get your AppKey and AppSecret from the ShadowBot Console API settings.
2. Input your credentials and Organization ID into the Connector configuration.
3. Ask your agent to list robots or start tasks to begin.

## Frequently Asked Questions

**What is the ShadowBot MCP?**
It is a way to connect your ShadowBot RPA robots to your AI agent. This allows you to manage, trigger, and monitor your automations directly through a chat interface.

**Can I start a robot task with this?**
Yes, you can trigger any ShadowBot RPA application using natural language commands. Your agent will handle the request and start the workflow for you.

**How do I see if my robots are working?**
You can ask your agent to list online robots or check specific task details. It will provide real-time updates on which bots are active and what they are doing.

**Can I manage my team's departments?**
Yes, the Connector allows you to view organizational departments and list their members. This makes it easy to manage team access and structure programmatically.

**How do I see if a task failed?**
Your agent can pull the logs for any specific task to show you the exact error. This helps you identify where a bot hit a snag without opening the full console.

**Is this for building new bots?**
No, this is for managing the bots you've already built in the ShadowBot platform. It provides the controls to run and monitor them, rather than writing the automation scripts.

**Do I need to keep the console open?**
No, you can manage everything through your AI client once the Connector is connected. You only need to visit the console to get your initial credentials.

**Can I automatically trigger an RPA task via the AI agent?**
Yes! Use the `trigger_task` tool with the target Robot ID and any required input parameters. Your agent will start the automation job and return the unique Job ID.

**How do I check if a robot is currently online and ready?**
Use the `list_robots` tool. It will retrieve all configured robots in your ShadowBot account along with their current status (e.g., Online, Busy, Offline).

**Can I retrieve the logs for a completed automation job?**
Yes! The `get_task_logs` tool allows you to fetch the execution details and any output data for a specific Job ID.