Skip to main content

​
Overview

Initialize a new sift-kg project in the current directory. Creates configuration files to get started with document extraction and knowledge graph building.

​
Usage

sift init [OPTIONS]

​
Options

​
--domain
string
Path to custom domain YAML to set in project config. If provided, this domain path will be written to the generated sift.yaml file.

​
Behavior

The init command creates two configuration files:
  1. .env.example - Template for environment variables including API keys
  2. sift.yaml - Project configuration file for domain, model, and extraction settings
If these files already exist, you will be prompted to confirm overwriting them.

​
Output Files

​
.env.example

Contains template for:
  • SIFT_OPENAI_API_KEY - OpenAI API key
  • SIFT_ANTHROPIC_API_KEY - Anthropic API key
  • SIFT_DEFAULT_MODEL - Default LLM model (e.g., openai/gpt-4o-mini)

​
sift.yaml

Contains project settings:
  • domain - Path to domain configuration or bundled domain name
  • model - Default LLM model
  • output - Output directory path
  • extraction - Extraction backend and OCR settings

​
Examples

​
Basic initialization

sift init
Creates default configuration files in the current directory.

​
Initialize with custom domain

sift init --domain path/to/custom-domain.yaml
Creates configuration files with the specified domain path in sift.yaml.

​
Next Steps

After running init, the command displays next steps:
  1. Copy .env.example to .env: cp .env.example .env
  2. Add your API key to .env
  3. (Optional) Edit sift.yaml to configure domain and settings
  4. Run extraction: sift extract ./docs/

​
See Also

  • domains - List available bundled domains
  • extract - Extract entities from documents

Build docs developers (and LLMs) love

Get started for freeTalk to us