# Chattermill MCP for AI Agents MCP

> Chattermill MCP lets your AI agent pull customer feedback, sentiment scores, and calculated metrics from dozens of sources—all in one place. Stop juggling dashboards; track themes, monitor NPS, and instantly understand what customers are actually saying across Zendesk, App Store, and more.

## Overview
- **Category:** customer-support
- **Price:** Free
- **Tags:** sentiment-analysis, nps, csat, feedback-loop, customer-experience, data-unification

## Description

This connector gives your AI client full control over your customer experience (CX) intelligence without forcing you to open a dashboard. You can connect the entire history of feedback—from support tickets to app reviews—and treat it like one unified data stream.

It goes beyond simple retrieval. Your agent analyzes the text for recurring topics, automatically grouping issues into actionable themes and categories. Need to know if things are getting better or worse? It calculates core metrics like NPS and CSAT on demand, allowing you to query them in plain conversation. If your current workflow involves manually cross-referencing data from Zendesk, Typeform, and App Store reviews just to get a mood report, this changes that. You're not limited to the tools inside one application; by connecting through Vinkius, you give your agent access to thousands of MCPs across industries. Just ask for the metrics or themes, and your AI client handles the complex data aggregation.

## Tools

### submit_feedback_response
Sends new customer feedback to a specific project, allowing you to tag it with a score or source type.

### get_chattermill_metric
Retrieves calculated metrics like NPS, CSAT, net sentiment, or overall volume for a given time frame.

### get_chattermill_project
Fetches specific details about a single customer feedback project by its unique ID.

### get_response_details
Provides full information for one piece of feedback, including the comment text and all applied themes.

### list_theme_categories
Lists high-level categories that group together related customer feedback topics for broad trend analysis.

### list_feedback_sources
Returns a list of all data sources connected to the project, such as Zendesk or App Store, helping you understand your data scope.

### list_data_types
Lists the specific types of classification used for responses, like NPS or review, aiding targeted filtering.

### list_chattermill_projects
First, you use this to list every available feedback project key in your entire Chattermill account.

### list_feedback_responses
Lists paginated customer responses for a specific project, allowing filtering by date range or source.

### list_custom_segments
Retrieves user-defined groups of customers for deep cohort analysis and advanced filtering.

### list_feedback_themes
Lists specific themes automatically generated by the system to classify recurring customer topics found in the feedback text.

## Prompt Examples

**Prompt:** 
```
What was our average satisfaction score last month?
```

**Response:** 
```
**Project:** Acme Corp | **Metric:** Average Score

| Period | Metric Type | Value |
| :--- | :--- | :--- |
| Last 30 Days | Average Score | 8.1/10 |
| Previous Month | Average Score | 7.5/10 |

*Analysis: The score increased by 0.6 points last month, suggesting recent updates helped.*
```

**Prompt:** 
```
Show me the biggest complaint themes from App Store reviews.
```

**Response:** 
```
**Top 3 Themes (App Store)**

1. **'Login Issues':** Volume: 45. Sentiment: Negative.
2. **'Slow Loading':** Volume: 38. Sentiment: Mixed.
3. **'Billing Confusion':** Volume: 22. Sentiment: Negative.

You can now filter by date or see details on a specific theme.
```

**Prompt:** 
```
What were the key issues for my mobile project?
```

**Response:** 
```
We found 8 active themes across your 'Mobile App' project, grouped under three major categories:

*   🚀 **Performance:** (e.g., Slow Loading, Crashes)
*   👤 **User Experience (UX):** (e.g., Navigation Issues, Confusing UI)
*   💳 **Billing/Payment:** (e.g., Invoice Errors, Subscription Confusion)

You can ask me to pull the volume for any of these categories.
```

## Capabilities

### List all configured feedback projects
You can identify every distinct customer feedback project set up in your Chattermill account.

### Fetch calculated sentiment metrics (NPS, CSAT, volume)
The agent retrieves specific performance scores, such as Net Promoter Score or average satisfaction ratings, for a given time period and project.

### Retrieve detailed theme analysis
You can browse AI-generated themes and categories to pinpoint recurring customer issues based on the text content.

### List available data sources and types
The agent shows you all connected data feeds, like Zendesk or App Store, so you know exactly what feedback is being analyzed.

### Get details on a single comment response
You retrieve all metadata for one piece of feedback, including the score and which themes were applied by the system.

### Submit new customer feedback entries
Your agent can send fresh feedback responses directly into a Chattermill project for immediate analysis.

## Use Cases

### The Quarterly Review Prep
A CX Manager needs to prepare a board presentation on customer mood. They ask the agent to fetch metrics, then list custom segments for 'Premium Users' who opened tickets last month. The agent compiles the NPS and sentiment volume by segment in minutes.

### Prioritizing Next Sprint Features
A Product Manager wants to know what feature is causing the most pain. They ask the agent to list feedback themes and filter for 'negative' sentiment within the last 30 days, instantly identifying the top three technical issues.

### Auditing Data Integrity
An Operations Team member suspects a data source is broken. They use list_feedback_sources to check connectivity and then run list_data_types to verify that all expected classifications are active.

### Real-Time Support Issue Triage
A support agent receives a critical piece of feedback via Typeform. Instead of filing it manually, they use submit_feedback_response right through the chat interface for immediate thematic analysis.

## Benefits

- Stop manually calculating metrics. Use get_chattermill_metric to retrieve live NPS, CSAT, or net sentiment scores instantly.
- Pinpoint exact product issues by asking for themes: list_feedback_themes shows you what customers are complaining about right now.
- Never lose context again. Get full details on a single comment response using get_response_details—you see the score, source, and themes all at once.
- Understand your data scope immediately. Use list_feedback_sources to verify if Zendesk or App Store data is actually feeding into your analysis.
- Run deep analyses without clicking tabs. List_custom_segments lets you create complex filter groups for cohort comparisons directly through conversation.

## How It Works

The bottom line is that instead of navigating multiple dashboards, you talk directly to your customer intelligence and get actionable metrics in return.

1. First, subscribe to this MCP and enter your unique Chattermill API Key from the dashboard.
2. Next, tell your AI client which customer feedback projects you want to analyze by listing them first.
3. Finally, ask your agent for specific insights—like 'What was our NPS last month?' or 'List all themes related to login issues'—to get the data back.

## Frequently Asked Questions

**How do I get a unified view of customer feedback using Chattermill MCP?**
You unify feedback by connecting your agent to this MCP. It pulls data from multiple sources like Zendesk and App Store, treating them as one continuous stream of insights rather than separate databases.

**Can I track NPS scores using Chattermill MCP for AI Agents?**
Yes, the agent retrieves calculated metrics, including Net Promoter Score (NPS) and CSAT. You can query these historical or current scores simply by asking a question.

**What if I need to analyze themes that aren't already defined?**
The MCP allows you to list existing themes, but it also lets your agent help categorize and find recurring topics in the raw text so you can prioritize features based on real pain points.

**Is Chattermill MCP for AI Agents good for product managers?**
It's excellent. Product Managers use it to identify patterns by asking for theme volumes, which helps them prioritize the roadmap using hard data instead of gut feeling.

**Does this MCP connect only to one type of feedback source?**
No. It is designed to unify feedback from dozens of channels—including support platforms and app stores—meaning your analysis covers your entire customer journey.