DEV Community: Emmanuel Mireku Mensah

getqueryly: MCP Server and API for AI Data Analysis (CSV to Charts in Minute)

Emmanuel Mireku Mensah — Tue, 15 Sep 2026 17:02:04 +0000

GetQueryly has an MCP server and API for AI data analysis. Upload a CSV, Excel, JSON, PDF, or Parquet file, ask in plain English, and get charts with short explanations plus a Quick Insight summary. Works out of the box with Claude, Cursor, and any MCP client, or via straight REST from your own app.

Docs: https://getqueryly.com/api-docs
Product: https://getqueryly.com

Why an MCP Server for Data Analysis?

AI assistants often guess about your data because they cannot see it directly. Point GetQueryly at your file, and the guesswork stops. The assistant calls a tool, the calculations run on your actual upload, and every result ships with a chart that proves it.
Use it when you want average revenue by region with visual proof, outliers flagged before you chart, or a consensus check from three distinct analyst passes on a hard call.
Setup: MCP in 5 Minutes
Endpoint: https://getqueryly.com/mcp/sse

Create an account at getqueryly.com.
Open Dashboard > API Keys > Generate key.
Ensure MCP is enabled on that key (toggled per key).
Add the following to your MCP client config:
{
"mcpServers": {
"getqueryly": {
"url": "https://getqueryly.com/mcp/sse",
"headers": {
"X-API-Key": "dmk_your_key_here"
}
}
}
}
Restart your client and ask "list getqueryly tools" to verify the connection.
Setup: REST API for Servers & Scripts
Same key, same costs. Send your key via the X-API-Key header. Base URL: https://getqueryly.com
1. Upload a File Max file size: 50MB. Supported types: CSV, XLSX, XLS, JSON, PDF, TXT, MD, PARQUET. curl -X POST https://getqueryly.com/api/data-scientist/upload \ -F "session_id=ds_1720000000_abc12" \ -F "file=@sales_data.csv"

Query the Session
curl -X POST https://getqueryly.com/api/data-scientist/query \
-F "session_id=ds_1720000000_abc12" \
-F "query=Show average revenue by region"
Health Checks, Insights & Chat

Health check

curl -H "X-API-Key: dmk_your_key_here" \
https://getqueryly.com/api/data-scientist/health/ds_1720000000_abc12

Insights

curl -H "X-API-Key: dmk_your_key_here" \
https://getqueryly.com/api/data-scientist/insights/ds_1720000000_abc12

Chat

curl -X POST https://getqueryly.com/api/chat \
-H "X-API-Key: dmk_your_key_here" \
-H "Content-Type: application/json" \
-d '{"message":"Summarize this session for standup"}'

Check Credit Limits ("Makes") curl -H "X-API-Key: dmk_your_key_here" \ https://getqueryly.com/api/makes/limits

Tool Costs & Pricing Model

Upload: Free
Chat: 1 Make
Data Query: 2 Makes
Health Check: 1 Make
Insights: 2 Makes
Swarm Quick: 4 Makes
Swarm Full: 6 Makes
Report: 1 Make > Pro-Tip: Upload once and reuse the same session_id for every follow-up to save credits. Check your limits before running a swarm analysis. Failed runs auto-refund so your balance is protected. Free tier covers daily testing, and paid packs stack. > 3 Workflows Worth Copying
1. Health Check Before Charts > Prompt: "Upload ./sales.csv with GetQueryly, run ds_health_check, and report missing values, outliers, and bad types." > Result: You get a data quality score out of 100 alongside per-column missing rates. Never chart dirty data again. >
2. Ask with Proof > Prompt: "Run ds_query on that session for average revenue by region, output a bar chart, and include a one-line summary under each bar." > Result: Returns precise numbers, visual charts, and ready-to-share summaries for Slack. >
3. Consensus for Hard Calls > Prompt: "Run ds_swarm_quick on churn drivers, then list where the analysts agree." > Result: Synthesizes independent analytical passes to catch nuances a single pass might miss. > FAQ Is GetQueryly an AI data analysis API? Yes. Upload, query, check health, generate insights, chat, and export over REST using a single header. View full details in the API Docs. Does it work as an MCP server for Claude and Cursor? Yes. Add the SSE URL and your API key to your client config. Exposed tools include ds_upload, ds_query, ds_health_check, ds_insights, ds_swarm_quick, ds_swarm_full, ds_report, ai_chat, chart, and more. What file formats are supported? CSV, Excel (XLSX/XLS), JSON, PDF, TXT, Markdown, Parquet, and images containing text up to 50MB per upload over REST.

Can I share results without login credentials?
Yes. One click generates a public share link containing the charts and narrative. Viewers do not need an account, while raw file downloads remain owner-only.

Where do I start?
Head over to getqueryly.com, grab an API key, check the API Docs, and run your first upload curl.

I built GetQueryly and will be hanging out in the comments! Post your file type and question below, and I'll reply with the exact prompt and cheapest tool chain to use.

GetQueryly: Ask Your Data Any Question. Get Real Answers.

Emmanuel Mireku Mensah — Tue, 11 Aug 2026 06:29:47 +0000

You have a spreadsheet. You have questions. You don't want to write code.

GetQueryly lets you ask your data anything in plain English and get real, actionable insights with charts, statistics, and reports.

What You Can Do
Upload any file. CSV, Excel, PDF, JSON, Parquet, SPSS - if it has data, we handle it.

Ask in plain English. No SQL. No formulas. No coding. Just type what you want to know.

Get real results. Actual charts. Actual statistics. Actual insights.

Features Built for Real Work
Data Health Check
Upload a file and instantly see what's wrong with it. Missing values, duplicates, outliers, correlations - everything flagged before you even ask.

Smart Analysis
Ask "show me revenue trends by region" and get a real chart with actual numbers. Ask "what's driving customer churn?" and get a statistical breakdown.

Multi-Analyst Reports
Need a full picture? GetQueryly assembles a team of AI analysts - each with different expertise. They collaborate and deliver a comprehensive report.

Anomaly Detection
Your data has problems hiding in plain sight. GetQueryly finds them automatically. Spikes, drops, outliers, unusual patterns - flagged and explained.

Data Storytelling
Numbers are boring. GetQueryly turns your data into a narrative. "Revenue grew 23% in Q3, driven primarily by the West region, with December being the strongest month."

Kaggle Dataset Explorer
200,000+ public datasets at your fingertips. Search by keyword, preview the data before loading, and analyze instantly. No downloads. No uploads. Just search and go.

Web Search Data Generation
Don't have data? Type a topic. GetQueryly searches the web across multiple sources, shows you what it found, and lets you generate a structured dataset from the research. Preview sources before committing.

Predictive Follow-Ups
After every analysis, GetQueryly suggests what to explore next. Not generic suggestions - real questions based on what your data actually shows.

Export Everything
PDF reports. Excel workbooks with multiple sheets. CSV data files. Chart images. Download in any format.

For Teams
Create a workspace. Share datasets. Collaborate on analysis. Manage who has access to what.

For Developers
Full REST API. MCP integration. Build your own tools on top of GetQueryly's analysis engine.

Free to Start
10 analyses per day. No credit card. No signup required to explore.

Try GetQueryly now →

GetQueryly - Your data has stories. We help you tell them.

Built an AI Document Assistant That Runs Code in a Sandbox

Emmanuel Mireku Mensah — Sat, 25 Jul 2026 14:51:08 +0000

Every AI document tool falls into one of two categories:

Chatbots that can talk about your files but can't download anything
Dumb converters that can merge PDFs but can't think
I built DocMake to bridge both worlds.

What DocMake Does
DocMake is an AI-powered document assistant at docmake.online. You upload a file, describe what you need in plain English, and AI picks the right tool to handle it.

The Cool Part: Sandboxed Code Execution
The AI Data Scientist feature is what I'm most proud of. Upload a spreadsheet and:

AI analyzes the data structure
Writes code to process it
Runs the code in a sandboxed subprocess (not in the main process)
Returns interactive charts + natural language insights
The key architectural decision: DataFrames never live in the main process. Every analysis runs in a forked subprocess with a timeout and memory limit. This prevents a single bad query from crashing the server.

The Analyst Swarm
Inspired by multi-agent systems, the Analyst Swarm assigns multiple AI personas to independently analyze your data. Each persona writes their own code, runs it, and reaches a consensus. It's like having a team of analysts working in parallel.

MCP Integration
DocMake is also an MCP (Model Context Protocol) server. This means AI assistants like Claude, Cursor, and VS Code can call DocMake's tools directly — not just talk about converting files, but actually do it.