GetQueryly — AI Data Analyst
Upload spreadsheets, get plain-English analysis with real computed charts and quality scores.
Installation
Ajoutez ceci à la configuration MCP de votre client (Claude Desktop : claude_desktop_config.json ; Cursor : .cursor/mcp.json ; VS Code et les autres utilisent la même forme « mcpServers »).
claude mcp add --transport http getqueryly-ai-data-analyst 'https://getqueryly.com/mcp/mcp' --header 'X-API-Key: <GetQueryly API key (free tier: 10 Makes/day)>'{
"mcpServers": {
"getqueryly-ai-data-analyst": {
"type": "http",
"url": "https://getqueryly.com/mcp/mcp",
"headers": {
"X-API-Key": "<GetQueryly API key (free tier: 10 Makes/day)>"
}
}
}
}Documentation
Tirée du README du projet sur GitHub, actualisée chaque semaine. Les liens ouvrent GitHub.
Sommaire
GetQueryly MCP Server
AI data analysis for your AI assistant. Connect Claude Desktop, Cursor, or any MCP client to GetQueryly and analyze spreadsheets in plain English — CSV, Excel, JSON, PDF, SPSS, images. The AI writes and runs real Python against your file, proves every number, and returns charts.
Hosted server — no install, no setup. Just point your client at the URL and add your API key.
Connect in 30 seconds
Claude Desktop / Cursor / any MCP client (streamable HTTP):
{
"mcpServers": {
"getqueryly": {
"url": "https://getqueryly.com/mcp/mcp",
"headers": { "X-API-Key": "YOUR_GETQUERYLY_API_KEY" }
}
}
}Get a free API key (10 free Makes/day): **https://getqueryly.com/keys**
What your assistant can do
| Tool | What it does |
|---|---|
analyze_file | One call deep analysis: upload → quality score → insights → chart → plain-English answer |
ds_upload | Upload CSV/Excel/JSON/PDF/SPSS/image, get a session id |
ds_query | Ask anything about your data — AI writes and runs real code |
ds_health_check | Honest 0-100 data quality score with exact issues |
ds_insights | Auto-surface patterns, anomalies, correlations |
ds_swarm | Multiple AI analyst personas analyze independently, then reach consensus |
chart | Bar, line, scatter, heatmap, box plot, treemap — any chart |
smart_test | AI picks the right statistical test and runs it |
ds_report | Generate a shareable PDF report |
ds_dashboard | Auto-build an executive dashboard (KPIs, charts) |
ds_graph | Knowledge-graph (GraphRAG) over your dataset |
kaggle_search / kaggle_load / kaggle_analyze | Search 200K+ Kaggle datasets and analyze them in place |
generate_diagram / create_infographic / create_comparison | Visual explainers from descriptions |
get_usage / get_costs | Free balance + pricing lookups |
Pricing
Free: 10 Makes/day with an API key (no card). Paid packs (GHS): Starter 50 / Growth 200 / Pro 600 — never expire. Full table: https://getqueryly.com/pricing
Links
- Web app: https://getqueryly.com
- API docs: https://getqueryly.com/api-docs
- llms.txt (for AI crawlers): https://getqueryly.com/llms.txt
- Keys: https://getqueryly.com/keys
Notes
- Transport: streamable HTTP (MCP 2025-03-26)
- Your files are processed in-memory; sessions auto-expire
- Every analysis is metered in "Makes" so costs are predictable
MIT licensed. Built by GetQueryly.
Official MCP Registry
Listed in the official registry: io.github.thesinbin12-jpg/getqueryly
Questions fréquentes
Qu'est-ce que GetQueryly — AI Data Analyst ?
Upload spreadsheets, get plain-English analysis with real computed charts and quality scores.
Comment installer GetQueryly — AI Data Analyst ?
Il fonctionne comme serveur distant. Dans Claude Code : claude mcp add --transport http getqueryly-ai-data-analyst 'https://getqueryly.com/mcp/mcp' --header 'X-API-Key: <GetQueryly API key (free tier: 10 Makes/day)>'. Pour les autres clients, collez la configuration JSON de cette page dans leurs réglages MCP.
Faut-il une clé API pour GetQueryly — AI Data Analyst ?
Oui. Il attend X-API-Key, à renseigner dans sa configuration.
GetQueryly — AI Data Analyst est-il open source ?
Oui. Son code est sur GitHub sous licence MIT.
Qui maintient GetQueryly — AI Data Analyst ?
Il est publié par thesinbin12-jpg. AI.tn le référence sans le maintenir ni le recommander.
Issu du registre MCP officiel. Être listé n'est pas une recommandation : vérifiez le code et les permissions d'un serveur avant de lui donner accès à vos données.