duckdb-analytics

Name: duckdb-analytics
Availability: InStock
Author: alex1980alex

Analyze JSONL and Parquet logs directly with DuckDB.

Updated Mar 29, 2026

Search Tags:#analytics #sql #latency #etl #parquet #duckdb #jsonl

Ready to run

npx skills add https://github.com/Alex1980Alex/1C-Framework --skill duckdb-analytics-alex1980alex

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: duckdb-analytics
Source: https://github.com/Alex1980Alex/1C-Framework/tree/main/.claude/skills/duckdb-analytics
Command: npx skills add https://github.com/Alex1980Alex/1C-Framework --skill duckdb-analytics-alex1980alex

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the need to migrate logs into a database just to answer analytical questions, letting you query JSONL and Parquet data directly with DuckDB.

Core Features & Use Cases

Direct log analytics: Read newline-delimited JSON and Parquet files in place for fast ad hoc analysis.
Latency and metric reporting: Compute p50, p95, p99, counts, and daily aggregates for operational logs.
Schema drift handling: Safely combine mixed log files with union-by-name reading and pre-cleaning patterns.
Export and archiving: Convert hot JSONL logs into Parquet for efficient cold storage and future reporting.

Quick Start

Ask the assistant to load a JSONL log with DuckDB, compute latency percentiles and daily aggregates, and export the results to Parquet.

Frequently Asked Questions about duckdb-analytics

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I analyze JSONL logs without loading them into a database?▼

You can analyze JSONL logs directly in place using DuckDB's read_json_auto function, which queries newline-delimited JSON files instantly without requiring any data migration or database setup.

What is the best way to compute p50, p95, and p99 latency percentiles from operational logs?▼

Computing latency percentiles from operational logs is done using DuckDB's approximate quantiles on JSONL or Parquet files, allowing you to calculate p50, p95, p99, and daily aggregates efficiently.

How do I query JSONL files that have inconsistent schemas or schema drift?▼

To query JSONL files with schema drift, DuckDB uses union_by_name reading alongside pre-cleaning patterns to safely combine mixed log files with varying structures into a single query result.

Can I query both hot JSONL and cold Parquet log tiers together in a single SQL statement?▼

Yes, you can query hot JSONL and cold Parquet log tiers together using DuckDB's in-process analytics engine, which applies parameterized SQL to read both file formats simultaneously without moving data.

How do I convert JSONL logs to Parquet for cold storage and reporting?▼

Converting JSONL logs to Parquet for cold storage is accomplished through DuckDB's Parquet export functionality, transforming hot newline-delimited JSON into archived Parquet files for efficient future reporting.

Do I need to install external dependencies to run SQL analytics on local Parquet and JSONL files?▼

No external dependencies are required to run SQL analytics on local Parquet and JSONL files, as DuckDB operates as a self-contained in-process database engine that reads these formats directly without external integrations.