# Apify MCP

> Apify MCP lets your agent audit and manage all cloud automation and web scraping jobs without logging into the external console. You can list every active automation setup, check recent runs for success rates, or pull specific data items from any dataset—all through natural chat conversation. It gives you real-time operational visibility across complex data pipelines.

## Overview
- **Category:** friends-mcp
- **Price:** Free
- **Tags:** automation, data-extraction, cloud-actors, task-scheduling, dataset-management, headless-browser

## Description

Okay, so if your team relies on automated scraping, managing that infrastructure is a pain. You're constantly checking logs and auditing results in a separate web console just to know if things are running right. This MCP changes that. It gives your AI agent direct access to the full scope of your automation ecosystem. Instead of clicking through complex menus, you just ask it questions about your jobs and data. Your agent handles the rest: it can list every single setup you have, check how recently a job ran, or pull specific records from massive datasets instantly. Because Vinkius hosts this MCP, you connect once to your preferred AI client and get immediate operational control over all your cloud automation tools. It means you spend less time checking dashboards and more time using the data.

## Tools

### list_webhooks
Lists configured webhooks that trigger actions when certain events occur.

### get_actor
Gets detailed information for a single, specific automation setup (actor).

### get_dataset_items
Retrieves individual data records from an established dataset.

### get_user_info
Provides basic details about the user connected to the system.

### list_actors
Lists all existing automation setups (actors) in your account.

### list_datasets
Lists all saved datasets containing scraped or generated data.

### list_key_value_stores
Retrieves a list of key-value storage containers used by your automation jobs.

### list_runs
Provides a history of execution runs for any specific actor or task.

### list_schedules
Lists the automated timing schedules configured to run your tasks periodically.

### list_tasks
Shows all defined, granular tasks within a specific automation setup.

## Prompt Examples

**Prompt:** 
```
List all my Apify actors.
```

**Response:** 
```
I've retrieved your actors. You have 5 active actors, including 'Website Content Crawler' and 'Google Maps Scraper'. Which one would you like to audit for recent runs?
```

**Prompt:** 
```
Show me the last 5 runs for actor ID xxxx.
```

**Response:** 
```
I've found the last 5 runs. The most recent one was completed successfully in 2 minutes. Two other runs had warnings. Would you like the full log details?
```

**Prompt:** 
```
Get items from dataset ID yyyy.
```

**Response:** 
```
I've retrieved the data from that dataset. It contains 50 items related to 'Competitor Pricing'. I can summarize the top findings or provide the full list.
```

## Capabilities

### Inventory Automation Setups
Lists every active web scraping or data pipeline setup configured in your account.

### Audit Task Runs and Schedules
Checks the history of automated tasks, showing when they ran, if they succeeded, and what their current schedules are.

### Retrieve Scraped Data Items
Pulls specific records or data points from your completed datasets for immediate review.

### Monitor Automation Components
Gathers metadata about the foundational pieces of your system, like webhooks and key-value stores.

## Use Cases

### Debugging Data Flow
A data scientist needs to know why yesterday's competitor pricing dataset is incomplete. They ask their agent, which then uses `list_runs` to find the last successful execution and `get_actor` to check that specific actor’s details for potential errors.

### Auditing Operational Health
An ops manager needs a quick report on all scheduled jobs. The agent uses `list_schedules` and then checks associated webhooks to confirm every automated process is set up with the proper alerts and triggers.

### Quick Data Validation
A growth lead just ran a major scrape and needs to see if it captured any data on 'Product X'. They ask the agent, which uses `list_datasets` followed by `get_dataset_items` to confirm the presence and quality of the specific records.

### Onboarding New Systems
A new team member needs to see what automation tools are available. They ask the agent to run `list_actors`, getting an immediate, comprehensive list of all running services without needing manual console access.

## Benefits

- Stop manual logging: Instead of navigating to a console, you can ask the agent to list all actors or audit tasks directly in your chat interface. It just works.
- Deep visibility into data: You don't have to download massive files to check a few records. The agent uses `get_dataset_items` to pull specific pieces of scraped information instantly.
- Verify job status quickly: Need to know if the scraping ran yesterday? Use `list_runs` and tell your agent which actor you need to check, getting immediate performance reports.
- Understand dependencies: By listing webhooks or key-value stores, you can map out exactly how different parts of your automation workflow are connected without guessing.
- Maintain control: You get a clear view of all operational components—schedules and tasks—using `list_schedules` and `list_tasks`, so nothing falls through the cracks.

## How It Works

The bottom line is: you control complex automation workflows using plain conversation instead of technical consoles.

1. First, subscribe to this MCP on Vinkius and input your Apify API Key.
2. Next, talk to your AI client. You just need to tell it what you want to check—for example, 'Show me the status of the pricing crawler.'
3. The agent calls the necessary tools, retrieves the data (like listing runs or getting dataset items), and presents a clean summary back in the chat.

## Frequently Asked Questions

**How does `list_actors` help me?**
It provides an immediate list of every automation setup in your account. This lets you quickly see what services are running and which ones might be outdated or unused.

**What is the difference between `list_runs` and `list_schedules`?**
`list_runs` shows actual historical job executions—what happened. `list_schedules` shows the rules for when jobs are set to run, so you know when they're *supposed* to fire.

**Do I need to use `get_dataset_items` every time?**
No. You only use it when you need to audit or retrieve specific data points from a completed dataset, rather than just checking the count of records in that dataset.

**`list_tasks` and `get_actor`—what's the relationship?**
You first use `get_actor` to get details on a whole setup. Then, you can use `list_tasks` to see the specific, granular steps that make up that overall automation process.

**When I use `get_user_info`, what details can I retrieve about my account?**
It provides the core metadata for your authenticated Apify user. You get confirmation of the owner ID, API access limits, and overall billing status right away.

**If a job fails, how do I troubleshoot it using `list_runs`?**
The run list shows not only success or failure, but also the specific error code and timestamp. You can check these details to narrow down whether the issue was an API problem or a script bug.

**What kind of custom data can I manage with `list_key_value_stores`?**
This tool lets you list and audit your persistent, structured key-value data. You can track settings, configurations, or small amounts of application state that aren't part of the main datasets.

**If I check my external integrations with `list_webhooks`, what should I look out for?**
You can see a list of all configured webhooks and their current endpoint status. This helps you monitor if your automated triggers are pointing to the correct, active URLs.

**How do I find my Apify API Key?**
Log in to your [**Apify Console**](https://console.apify.com/account/integrations), and you will find your API Token under the **Integrations** tab. Copy and paste it below.

**Can the agent check the results of a scrape?**
Yes. Use the `get_dataset_items` tool providing the Dataset ID. Your agent will retrieve the items from the cloud storage, allowing you to audit the output instantly.

**Is it possible to list actor runs via the agent?**
Yes. The `list_runs` tool allows your agent to retrieve the history of executions for any specific actor, including durations and final statuses.