# Hightouch (Reverse ETL) MCP

> Hightouch (Reverse ETL) lets you manage data pipelines and synchronize data from your warehouse back out to SaaS tools entirely through conversation. You list all sync schedules, monitor historical runs for success rates, audit the underlying SQL models, and map every connected source or destination without ever opening a dashboard.

## Overview
- **Category:** growth-engine
- **Price:** Free
- **Tags:** reverse-etl, data-sync, warehouse-integration, data-modeling, pipeline-orchestration, saas-integration

## Description

Managing enterprise data flows means connecting Snowflake or BigQuery to external apps like Salesforce or HubSpot, and that process is complex. This MCP gives your agent full control over those Reverse ETL workflows using natural language. You don't need to click through multiple tabs or read error codes in a restrictive UI. Instead, you simply ask questions: "Show me the run history for sales data," or "Which SaaS apps are receiving objects from our warehouse?" Your agent answers immediately, providing detailed metrics and schemas on demand.

This ability to audit complex data models and sync performance via chat changes how quickly your team can validate data integrity. When you connect this MCP through Vinkius, you gain instant access to the entire catalog of tools needed for robust data operations right from your preferred AI client.

## Tools

### list_destinations
Lists every SaaS application node that receives synchronized objects from your warehouse.

### list_syncs
Retrieves a list of all current data synchronization schedules configured in Hightouch.

### get_sync
Fetches specific, detailed information about an individual sync job.

### list_sync_runs
Retrieves historical records of all data synchronization runs for performance monitoring.

### list_models
Provides a list and audit of the SQL definitions used in your analytics pipelines.

### list_sources
Enumerate all active internal data warehouse connections, such as Snowflake or BigQuery.

## Prompt Examples

**Prompt:** 
```
List all my current Hightouch syncs
```

**Response:** 
```
I've found 5 active syncs. Highlights include 'Snowflake to Salesforce' (ID: sync-001), 'BigQuery to HubSpot' (ID: sync-004), and 'Postgres to Slack' (ID: sync-007). Would you like the status of the most recent run for any of these?
```

**Prompt:** 
```
Show me the last 5 runs for sync ID 'sync-001'
```

**Response:** 
```
Retrieving run history for 'sync-001'… The last 5 runs all finished with a 'Success' status. A total of 12,450 rows were processed in the most recent run (finished 2 hours ago). Would you like to see if any errors were logged in previous weeks?
```

**Prompt:** 
```
Which data sources are connected to our Hightouch account?
```

**Response:** 
```
I've identified 3 connected data sources: 'Production Snowflake', 'Stripe BigQuery Dataset', and 'Internal Postgres'. All connections are showing as 'Active'. Would you like to see the available models for 'Production Snowflake'?
```

## Capabilities

### Audit Data Destinations
List every connected SaaS application (like Salesforce or HubSpot) that is currently receiving synchronized objects.

### Check Synchronization Schedules
Retrieve a list of all active data sync jobs and get details on their schedules and underlying schemas.

### Review Run Performance History
Extract detailed records of past execution states, allowing you to monitor overall reliability and success logs for any job.

### Inspect Data Models
Audit the SQL definitions used in your warehouse analytics pipelines that drive data synchronization.

### Map Internal Sources
Enumerate and securely map all active internal data warehouse connections, such as Snowflake or BigQuery.

## Use Cases

### Investigating a Missing Metric in Salesforce
The Growth Ops Manager notices sales numbers are stale in HubSpot. They ask their agent to check the run history for the specific sync, using list_sync_runs. The agent confirms the last 10 runs succeeded but points out that the underlying model (list_models) was recently updated and needs review.

### Onboarding a New Data Warehouse Connection
The Analytics Engineer needs to confirm if BigQuery is ready for outbound sync. They use list_sources to enumerate all active connections, confirming the schema map from the new platform before writing any data flows.

### Auditing Compliance and Data Flow
A compliance officer asks their agent: "Which apps receive PII?" The agent uses list_destinations to identify all connected SaaS nodes, providing an instant audit trail of where sensitive data is flowing.

### Troubleshooting a Broken Sync
The Data Engineer notices the 'Marketing' sync hasn't updated. They ask the agent for details on the specific job (get_sync) and find that while the sync was scheduled, the underlying data model definition needs adjustment.

## Benefits

- Stop manually opening dashboards. You can check run history and get success logs simply by asking your agent for it, saving significant time.
- Audit every connected destination—from Salesforce to Slack—using list_destinations. Know exactly where your synchronized data is going without guessing.
- When a sync fails, don't wait for alerts. Use the run history tools (list_sync_runs) to immediately see past execution states and diagnose performance issues.
- Verify data source connections instantly. You can list all sources using list_sources to confirm if your warehouse link is active and mapped correctly.
- Deeply inspect pipeline logic by listing models. This lets you audit the underlying SQL definitions that drive complex data transformations.

## How It Works

The bottom line is you manage complex data synchronization tasks by talking to it, not by clicking through dashboards.

1. Subscribe to this MCP and provide your Hightouch API Key.
2. Connect the key to any MCP-compatible client (Claude, Cursor, etc.).
3. Ask your agent specific questions about sync health or data models.

## Frequently Asked Questions

**How do I check if a specific data sync is running with Hightouch? (list_syncs)**
You use list_syncs to get an immediate overview of all active schedules. This lets you see which pipelines are currently scheduled or need attention across your entire account.

**What does the MCP help me audit about my data models? (list_models)**
It allows you to list and inspect the underlying SQL definitions that power your syncs. This is critical for knowing exactly how your raw warehouse data gets shaped before it leaves.

**Can I see if a destination like Salesforce is connected? (list_destinations)**
Yes, you use list_destinations to identify all connected SaaS nodes. This confirms that the endpoint is ready and receiving objects from your warehouse.

**How can I check if a data source connection is working? (list_sources)**
You ask the agent to list sources, which enumerates all active connections like Snowflake or BigQuery. This verifies that the initial internal link feeding your process is stable.

**If I have a sync error, how do I find out why? (list_sync_runs)**
Use list_sync_runs to pull historical execution states and success logs. This gives you the performance metrics needed to pinpoint exactly when and where an issue started.