Skip to main content
Glama
stackcurious

revenuecat-charts-mcp

by stackcurious
README.md
# @stackcurious/revenuecat-charts-mcp

An MCP server that gives AI agents access to RevenueCat's Charts & Metrics API. Query MRR, churn, trial conversion, revenue, and 17 more subscription metrics — directly from Claude, Cursor, or any MCP client.

RevenueCat's official MCP server has 26 tools for project configuration. This server fills the gap: **analytics**.

## Why

Your AI agent can configure RevenueCat projects via MCP. But it can't check if those changes are working. That's like giving an agent a steering wheel but no dashboard.

This server gives agents eyes on the data.

## Tools

| Tool | Description |
|------|-------------|
| `rc_get_overview` | Snapshot of MRR, active subs, trials, revenue, new customers, active users |
| `rc_get_chart` | Query any of 21 chart types with filters, segments, date ranges |
| `rc_get_chart_options` | Discover available filters, segments, and resolutions for a chart |
| `rc_health_check` | Comprehensive health report: MRR + churn trend + trial conversion + signals |
| `rc_compare_periods` | Compare any metric across two time periods with deltas and % changes |

### Available Charts

`actives` · `actives_movement` · `actives_new` · `arr` · `churn` · `cohort_explorer` · `conversion_to_paying` · `customers_new` · `customers_active` · `ltv_per_customer` · `ltv_per_paying_customer` · `mrr` · `mrr_movement` · `refund_rate` · `revenue` · `subscription_retention` · `subscription_status` · `trials` · `trials_movement` · `trials_new` · `trial_conversion_rate`

## Setup

### 1. Install

```bash
npm install -g @stackcurious/revenuecat-charts-mcp
```

### 2. Configure your MCP client

**Claude Code:**
```bash
claude mcp add revenuecat-charts -- env REVENUECAT_API_KEY=sk_your_key_here npx @stackcurious/revenuecat-charts-mcp
```

**Claude Desktop / Cursor / Windsurf** — add to your MCP config:
```json
{
  "mcpServers": {
    "revenuecat-charts": {
      "command": "npx",
      "args": ["@stackcurious/revenuecat-charts-mcp"],
      "env": {
        "REVENUECAT_API_KEY": "sk_your_key_here"
      }
    }
  }
}
```

### 3. Use it

Ask your AI agent:
- *"Give me a health check on my subscription app"*
- *"Show me my MRR trend for the last 3 months"*
- *"Compare my churn rate this month vs last month"*
- *"What's my trial conversion rate broken down by store?"*

## Example Output

### Health Check
```
# Subscription Health Report

*Generated 2026-04-03*

## Current Snapshot

**MRR**: $4,555
**Active Subscriptions**: 2,534
**Active Trials**: 77
**Revenue (28d)**: $4,712
**New Customers (28d)**: 1,516
**Active Users (28d)**: 13,702
**ARPU**: $1.80/mo
**Trial Pipeline**: 2.9% of active base in trial

## Churn Trend (Monthly)
**Current**: 6.98%
**Prior month**: 6.67%
**Trend**: increasing

## Health Signals
- Churn between 5-10% — monitor closely
- Trial conversion 25-50% — room to optimize
- Strong acquisition — new customers > 10% of base
```

### Period Comparison
```
# MRR — Period Comparison

**Period A**: 2026-02-01 → 2026-02-28
**Period B**: 2026-03-01 → 2026-03-31

**MRR**: $4,561.94 → $4,541.72 (↓ 0.4%)
```

## Rate Limiting

The RevenueCat Charts API has a rate limit of ~5 requests/minute. This server includes built-in rate limiting with automatic queuing — you don't need to worry about it.

## Requirements

- Node.js 18+
- A RevenueCat secret API key (`sk_...`) with Charts & Metrics permissions

## API Key

Get your API key from [RevenueCat Dashboard](https://app.revenuecat.com) → Project Settings → API Keys. Create a secret key with `charts_metrics:overview:read` and `charts_metrics:charts:read` permissions.

## Built by

[Q](https://getqpro.com) — an AI agent orchestration platform. This server was built as part of Q's application for [RevenueCat's Agentic AI Advocate role](https://getqpro.com/blog/revenuecat-application).

## License

MIT

TDQS

A3.7/5.0

Scored across 5 tools

Disambiguation3/5

Most tools are distinct, but rc_get_overview and rc_health_check overlap significantly: both surface MRR, active subscribers, trials, and revenue. The descriptions help separate a quick snapshot from a comprehensive health report, but an agent could easily pick the wrong one for a simple metrics request.

Naming Consistency4/5

Tools consistently use the rc_ prefix and snake_case, with clear verbs like get and compare. The pattern is mostly verb_noun, but rc_health_check is a noun phrase rather than verb_noun, and rc_get_chart_options / rc_get_chart are very similar in form, so minor inconsistency exists.

Tool Count5/5

Five tools is well-scoped for a RevenueCat charting/analytics MCP server. Each tool serves a distinct purpose in the workflow: discover options, query charts, get overviews, compare periods, and generate health reports.

Completeness4/5

The tool surface covers the core charting and analytics needs: listing chart types, querying time-series data, discovering chart parameters, comparing periods, and getting health summaries. A minor gap is the lack of a dedicated tool for metadata like available metrics or chart lists, though this is partially embedded in rc_get_chart and rc_get_chart_options.

Maintenance

ActivityInactive
ResponsivenessNo issues