How do I install the Dct Infer skill?

Run npx skillmds@latest add tomevault-io/dct-infer in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Dct Infer safe to use?

This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Dct Infer?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Dct Infer free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Dct Infer?

tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Dct Infer by tomevault-io — Agent Skill

Name: Dct Infer
Author: tomevault-io

Dct Infer

Use this skill when the user wants to generate SQL CREATE TABLE statements from data files, infer schema from CSV/JSON/Parquet, create database schemas from existing data, or get column types from a file. Triggers include "generate schema", "create table from csv", "infer types", "what's the schema", "get column types", "sql ddl", or when preparing data for SQL databases like DuckDB, PostgreSQL, or similar.

tomevault-io Updated Sep 17, 2026

File contents

DCT Infer - Generate SQL Schema

Create DuckDB-compatible CREATE TABLE statements by analyzing data file contents.

When to Use

Use this skill when you need to:

Create database tables from existing data files
Document the schema of a dataset
Generate DDL for ETL pipelines
Understand column types in a file
Prepare data for SQL-based analysis

Installation

which dct || go build -o dct && chmod +x ./dct

Usage

dct infer <file> [flags]

Flags

-t, --table <name>: Table name (default: "default")
-n, --lines <number>: Number of lines to analyze for type inference (useful for large files)
-o, --output <file>: Output to file instead of stdout

Examples

Basic schema inference:

dct infer data.csv

With custom table name:

dct infer data.parquet -t events

Save schema to file:

dct infer large.ndjson -n 1000 -t users -o schema.sql

Infer from specific number of rows:

dct infer bigfile.csv -n 500 -t transactions

Output Format

DuckDB-compatible CREATE TABLE statement:

create table users (
    "id" bigint,
    "name" varchar,
    "email" varchar,
    "created_at" timestamp,
    "is_active" boolean
)

Supported Data Types

The inferred schema uses DuckDB types:

bigint - 64-bit integers
integer - 32-bit integers
double - Floating point numbers
varchar - String/text data
timestamp - Date and time
date - Date only
time - Time only
boolean - True/false values
array(...) - Array columns
row(...) - Struct/nested columns

Best Practices

Use -n flag for large files to speed up inference
Column names are quoted to handle special characters
Output is compatible with DuckDB and similar SQL databases
For Parquet files, types are read directly from metadata
For CSV/JSON, types are inferred from sample data

Integration Examples

With DuckDB

# Create table directly
dct infer data.csv -t my_table | duckdb mydb.duckdb

# Or save and execute
dct infer data.csv -t my_table -o schema.sql
duckdb mydb.duckdb < schema.sql

In Scripts

#!/bin/bash
for file in *.csv; do
    dct infer "$file" -t "$(basename "$file" .csv)" > "${file%.csv}.sql"
done

Related Skills

dct-peek: Preview data before inferring schema
dct-profile: Check data quality before creating tables

Converted and distributed by TomeVault — claim your Tome and manage your conversions.

---name: dct-inferdescription: Use this skill when the user wants to generate SQL CREATE TABLE statements from data files, infer schema from CSV/JSON/Parquet, create database schemas from existing data, or get column types from a file. Triggers include "generate schema", "create table from csv", "infer types", "what's the schema", "get column types", "sql ddl", or when preparing data for SQL databases like DuckDB, PostgreSQL, or similar.---# DCT Infer - Generate SQL SchemaCreate DuckDB-compatible CREATE TABLE statements by analyzing data file contents.## When to UseUse this skill when you need to:- Create database tables from existing data files- Document the schema of a dataset- Generate DDL for ETL pipelines- Understand column types in a file- Prepare data for SQL-based analysis## Installation```bashwhich dct || go build -o dct && chmod +x ./dct```## Usage```bashdct infer <file> [flags]```## Flags- `-t, --table <name>`: Table name (default: "default")- `-n, --lines <number>`: Number of lines to analyze for type inference (useful for large files)- `-o, --output <file>`: Output to file instead of stdout## ExamplesBasic schema inference:```bashdct infer data.csv```With custom table name:```bashdct infer data.parquet -t events```Save schema to file:```bashdct infer large.ndjson -n 1000 -t users -o schema.sql```Infer from specific number of rows:```bashdct infer bigfile.csv -n 500 -t transactions```## Output FormatDuckDB-compatible CREATE TABLE statement:```sqlcreate table users ( "id" bigint, "name" varchar, "email" varchar, "created_at" timestamp, "is_active" boolean)```## Supported Data TypesThe inferred schema uses DuckDB types:- `bigint` - 64-bit integers- `integer` - 32-bit integers- `double` - Floating point numbers- `varchar` - String/text data- `timestamp` - Date and time- `date` - Date only- `time` - Time only- `boolean` - True/false values- `array(...)` - Array columns- `row(...)` - Struct/nested columns## Best Practices- Use `-n` flag for large files to speed up inference- Column names are quoted to handle special characters- Output is compatible with DuckDB and similar SQL databases- For Parquet files, types are read directly from metadata- For CSV/JSON, types are inferred from sample data## Integration Examples### With DuckDB```bash# Create table directlydct infer data.csv -t my_table | duckdb mydb.duckdb# Or save and executedct infer data.csv -t my_table -o schema.sqlduckdb mydb.duckdb < schema.sql```### In Scripts```bash#!/bin/bashfor file in *.csv; do dct infer "$file" -t "$(basename "$file" .csv)" > "${file%.csv}.sql"done```## Related Skills- `dct-peek`: Preview data before inferring schema- `dct-profile`: Check data quality before creating tables---> Converted and distributed by [TomeVault](https://tomevault.io/claim/andrew-a-hale) — claim your Tome and manage your conversions.

Frequently asked questions