# Hevo Data MCP

> Hevo Data (ETL & Data Pipeline) lets you manage your entire data integration stack using natural conversation. List pipelines, check destination status across BigQuery or Snowflake, and audit row usage without jumping between dashboards. Take full control of your automated ETL orchestration directly from your AI client.

## Overview
- **Category:** developer-tools
- **Price:** Free
- **Tags:** etl, data-pipelines, data-warehousing, data-integration, pipeline-monitoring, automated-sync

## Description

Managing complex data flows usually means opening five different tabs: one for pipeline status, another for billing metrics, a third to check if the data hit BigQuery, and so on. This MCP changes that by giving you direct conversational access to your Hevo Data account. You can ask your AI client simple questions—like 'Are my Snowflake destinations healthy?' or 'How many rows did I use this month?'—and get immediate answers. It lets you orchestrate pipelines and monitor every connection, from the transformation models defining your logic to the final billing usage report. If you're building your agent catalog on Vinkius, adding this MCP means your users can manage mission-critical data assets without ever leaving their chat window. You simply tell your AI client what you need, and it executes the checks across all your connected data destinations.

## Tools

### list_pipelines
Lists all active data pipelines currently running.

### get_pipeline
Retrieves specific details about a single pipeline.

### list_destinations
Shows you every connected data warehouse destination (e.g., BigQuery, Snowflake).

### list_models
Retrieves a list of all defined transformation models.

### list_workflows
Lists the complex workflows that connect multiple data transformations together.

### get_usage
Reports on your account's current usage metrics and billing limits.

## Prompt Examples

**Prompt:** 
```
List all my active Hevo pipelines
```

**Response:** 
```
I've found 6 active pipelines. Highlights include 'Stripe to BigQuery' (ID: pipe-001), 'HubSpot to Snowflake' (ID: pipe-005), and 'Shopify to Redshift' (ID: pipe-008). Would you like to check the usage for any of these?
```

**Prompt:** 
```
Show me the destinations for my 'Sales Data' pipeline
```

**Response:** 
```
The 'Sales Data' pipeline (ID: pipe-005) is currently replicating to 2 destinations: a BigQuery production dataset and a Snowflake staging warehouse. Both are showing 'Healthy' status. Would you like to see the last sync time?
```

**Prompt:** 
```
How much of my row quota have I used this month?
```

**Response:** 
```
You've used 12.5 million rows out of your 20 million row monthly quota (62.5%). At current ingestion rates, you are projected to finish the month at 18.2 million rows. I can break down usage by pipeline if needed.
```

## Capabilities

### List active pipelines
Retrieves a list of every automated ETL pipeline configured in your account.

### Check destination health
Analyzes the status and connection details for all data warehouses like BigQuery, Snowflake, or Redshift.

### Audit account usage
Pulls real-time metrics on your row replications and overall billing usage against your quota.

### List transformation models
Shows the specific mappings and logic attached to keep your data quality consistent.

### Discover workflow connections
Maps out complex, multi-step data workflows connecting different transformations across your stack.

## Use Cases

### The nightly sync failed; I need to know why.
An Ops Manager asks the agent: 'What's wrong with my data flow?' The agent calls get_pipeline and list_destinations, reporting that the pipeline is down because the Snowflake destination connection timed out. The manager fixes it immediately without logging into any web UI.

### I need to audit our billing before Q3.
An Analytics Lead queries: 'How many rows did I use this month?' The agent calls get_usage, providing a usage breakdown and projection. This prevents unexpected overages when the data team scales up reporting.

### We added a new staging area; where does it go?
A Data Engineer uses list_destinations to confirm that their new Redshift cluster is correctly recognized by Hevo. They then use get_pipeline to ensure the required sync job is configured for that specific target.

### I need to prove data lineage for an audit.
The agent uses list_workflows combined with list_models, generating a map of every transformation step and mapping. This proves exactly how raw source data becomes final report metrics for compliance checks.

## Benefits

- Stop jumping between dashboards. You can list pipelines and check destination status—all in one conversation.
- Get an instant audit of your account usage by calling get_usage, so you never exceed your row quota unexpectedly.
- Verify data integrity using list_models to track the exact mappings bounding your staging logic for quality assurance.
- Map out complex connections using list_workflows. You'll see how multiple transformations link across your entire data stack.
- Quickly understand which sources feed into which targets by listing all destinations, ensuring no critical warehouse is missed.

## How It Works

The bottom line is you manage complex, critical data infrastructure using only chat commands.

1. Subscribe to this MCP and provide your unique Hevo Data API key and region (e.g., US or EU).
2. Connect the credentials to your preferred AI client, like Cursor or Claude.
3. Ask a natural language question—for example, 'Check my data pipeline usage'—and get immediate, actionable reports.

## Frequently Asked Questions

**How does Hevo Data (ETL & Data Pipeline) MCP help with billing?**
You call get_usage to instantly check how many rows you've replicated and what your remaining quota is. This prevents unexpected overages by keeping usage metrics visible in the chat.

**Can I list all my pipelines using Hevo Data (ETL & Data Pipeline) MCP?**
Yes, calling list_pipelines gives you a full rundown of every automated ETL pipeline configured in your account right from the agent interface.

**Does this MCP work with Snowflake and BigQuery?**
It monitors destinations for major data warehouses like Snowflake, BigQuery, and Redshift. You can list_destinations to confirm connectivity across all these systems.

**What is the difference between get_pipeline and get_usage?**
get_pipeline gives specific details on a single data flow's configuration, while get_usage reports generalized account metrics like total row replication count and billing limits.

**Is this useful for checking my transformation logic?**
Yes. Use list_models to review the explicit mappings that define your staging data logic and ensure quality standards are met before reporting.