ETLai 0.3.5

Local, folder-driven Dagster CSV transformation engine with hot-folder sensors, AI-assisted pipeline creation, and Tkinter configuration UI.

ETLai

A local, AI-assisted data transformation engine built on Dagster. Install it, scaffold a project, and let Claude Code create new pipelines — from CSV transformations to API ingestion.

Install

pip install ETLai

Requires Python 3.10+ and Tkinter (ships with most Python installations).

Quick start

etlai init ~/my-etl
cd ~/my-etl
etlai sync
etlai run

This starts a Dagster dev server at http://localhost:3000. Enable sensors in the UI, then drop CSV files into pipelines/<name>/inbox/.

Commands

Command	Purpose
etlai init <dir>	Scaffold a new project with example pipelines
etlai sync	Validate manifests, create folders, prompt for path: ask, check env files
etlai run	Start the Dagster dev server
etlai list	Show all registered pipelines

How it works

Each pipeline is defined by a manifest (pipelines/<name>/manifest.yaml) that wires together:

• Atom — a reusable, single-unit-of-work transformation
• Form — first-run Tkinter UI that collects config (or passthrough for no-UI)
• Trigger — what causes the pipeline to run (file sensor, cron schedule, or both)

Trigger fires → load files (if any) → configure (form/config.json) → execute atom
                                        ├→ processed/ + output/     (success)
                                        └→ rejected/ + *.error.txt  (failure)

Pipeline types

File-based (CSV transformation)

Drop files into inbox/. Sensor detects stable files and triggers the pipeline.

name: vlookup_rollnumber
atom: vlookup
form: vlookup_column_picker
min_files: 2

API-based (data ingestion)

Fetch data from REST APIs on a schedule. No inbox files needed.

name: fetch_hr_data
atom: hr_api_fetch
form: passthrough
min_files: 0
env_file: ~/.etlai/secrets.env
requires_env:
  - HR_API_TOKEN
trigger:
  rules:
    - type: schedule
      cron: "0 8 * * *"

Composite (multi-step chains)

Chain multiple atoms. Output of each step feeds into the next.

name: vlookup_then_groupby
min_files: 2
steps:
  - atom: vlookup
    form: vlookup_column_picker
  - atom: groupby
    form: groupby_picker

Shipped atoms

Atom	Description
vlookup	Left join two CSVs on specified columns with dtype validation
groupby	Group by column with count, sorted descending
mock_generate	Generate synthetic data from file headers using Faker
api_fetch	Generic REST API fetcher (JSON/XML/CSV, field mapping, env-based auth)

Shipped forms

Form	Description
vlookup_column_picker	Join + output column multi-select with dtype validation
groupby_picker	Single column selection
passthrough	No UI — reads config.json and passes to atom

Key concepts

Reference folder

reference/ holds permanent data used across runs (lookup tables, previous API snapshots). The framework passes reference file paths to atoms automatically.

Custom data paths

path: ask in a manifest prompts the user (via folder picker) during etlai sync to choose where data folders live. Each pipeline can have its own location.

Credentials

API credentials live in env files outside the project (e.g. ~/.etlai/secrets.env). The framework loads them before atom execution. etlai sync validates required vars are present without printing values.

Adding pipelines with Claude Code

Open your project in Claude Code. The scaffolded CLAUDE.md teaches it to:

1. Create atoms (or reuse shipped ones)
2. Create forms (or use passthrough with pre-written config.json)
3. Write manifest.yaml
4. Run etlai sync

Resolution order

• Atoms: ./atoms/<name>.py → etlai.atoms.<name> (package)
• Forms: ./forms/<name>.py → etlai.forms.<name> (package)

User files take precedence over shipped ones.

Development

git clone <repo>
pip install -e .
etlai init /tmp/test-project
cd /tmp/test-project
etlai run