mlld:query

Generate and execute SQL or code queries across databases, CSVs, and files.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/mlld-lang/mlld --skill mlld-query
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mlld:query
Source: https://github.com/mlld-lang/mlld/tree/main/plugins/mlld/skills/query
Command: npx skills add https://github.com/mlld-lang/mlld --skill mlld-query

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates cross-source data analysis by letting an LLM write SQL or code to query data, then executes it locally with zero LLM tokens, and performs parallel deep research on the filtered results.

Core Features & Use Cases

  • LLM-assisted query generation across databases, CSVs, and documents
  • Deterministic code execution for fast, token-free data access
  • Parallel deep research and synthesized insights from filtered results
  • Use Case: When questions require complex cross-tabulation and slicing across sources

Quick Start

Ask a cross-source data question and let the skill generate and execute SQL or code queries across sources, then synthesize the results.

Frequently Asked Questions about mlld:query

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

FAQPage Schema
How do I run cross-source data analysis across databases and CSV files?

Cross-source data analysis is handled by letting an LLM write SQL or code to query databases and CSVs, then executing it locally without LLM tokens to filter and slice the combined results.

Can I use LLM-generated SQL to query local CSV files without spending tokens?

Yes, LLM-generated SQL queries local CSV files without spending tokens by executing the generated code deterministically on your machine, enabling fast data access and cross-tabulation.

What is the best way to perform parallel deep research on filtered data results?

Parallel deep research on filtered data results is orchestrated through a structured synthesis workflow that runs multiple investigations concurrently on the sliced outputs.

Does cross-source data discovery support read-only access for database querying?

Cross-source data discovery enforces strict read-only access for database querying, ensuring that LLM-generated SQL only retrieves data without risking modifications to the source.

How do I cross-tabulate and slice data across logs and documents?

Cross-tabulation and slicing across logs and documents is achieved by exposing the schema to an LLM, which then writes the code to filter the sources and apply result limits.

What are the limitations of using LLM-assisted query generation for data synthesis?

Limitations of LLM-assisted query generation include enforced result limits on the filtered data and strict read-only access, meaning you cannot write data back to the queried sources.