How do I install the Get Started skill?

Run npx skillmds@latest add keboola-ai-kit/get-started 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 Get Started safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: makes network calls, reads secrets. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Get Started?

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 Get Started free to use?

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

Who published Get Started?

keboola (@keboola-ai-kit) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Get Started by keboola-ai-kit — Agent Skill

Name: Get Started
Rating: 4.5 (4 reviews)
Author: keboola

Get Started

Mandatory initialization workflow for new Keboola Python components — use this skill before writing any code. Contains the exact cookiecutter command, Keboola-specific post-template cleanup steps, KBC_DATADIR directory structure, data/config.json format, and first commit format that are easy to get wrong without this reference. Invoke whenever the user is starting from a fresh or empty repository, has never built a Keboola component before, wants to run the cookiecutter template, or says things like "new extractor for X", "build ex-something from scratch", "fresh repo", "empty git repo", "brand new component", "never done this in keboola", "initialize a component", "scaffolding a new component", "just initialized the git repo", "nothing set up yet". Not for existing components that already have src/component.py.

keboola Updated Sep 10, 2026

File contents

Initialize New Keboola Component

Before writing any code, check whether the component has already been initialized.

Step 1: Detect state

Check for signs of an existing cookiecutter setup:

ls src/component.py component_config/ pyproject.toml 2>/dev/null

If all present → skip initialization, hand off to @develop-component immediately
If missing or empty repo → proceed with initialization below

Step 2: Gather component info

You may already know the component details from the conversation. If not, ask:

Component name — kebab-case, no type suffix (e.g. ex-salesforce, not salesforce-extractor)
Component ID — vendor.component-name format (e.g. keboola.ex-salesforce)
Type — extractor, writer, or application
Short description — one sentence of what it does

Naming rules:

Never include "extractor", "writer", or "application" in the name itself
Use prefix convention: ex- for extractors, wr- for writers

Step 3: Run cookiecutter

# Install if missing
which cookiecutter || pip install --user cookiecutter

cookiecutter gh:keboola/cookiecutter-python-component

Run interactively so the user can confirm the values. The template creates src/, component_config/, tests/, .github/workflows/, Dockerfile, pyproject.toml, and data/ with example files.

Step 4: Clean up and configure

# Remove the generic example files that come with the template
find data -type f -delete

# Recreate the expected directory structure
mkdir -p data/in/tables data/in/files data/out/tables data/out/files

Then create data/config.json with realistic example parameters based on what the component actually does — not a generic placeholder. Read component_config/configSchema.json if it already has content, and mirror the required fields with plausible example values. Use # prefix for sensitive fields (e.g. "#api_key").

Step 5: Initial commit

git add -A
git commit -m "feat: initialize component from cookiecutter template"

Step 6: Hand off

Once initialized, pass to @develop-component to start the implementation. Share what you know about the component's purpose so it has context.

Key resources

Cookiecutter template: gh:keboola/cookiecutter-python-component
Developer Portal: https://components.keboola.com/
Component tutorial: https://developers.keboola.com/extend/component/tutorial/

---name: get-starteddescription: Mandatory initialization workflow for new Keboola Python components — use this skill before writing any code. Contains the exact cookiecutter command, Keboola-specific post-template cleanup steps, KBC_DATADIR directory structure, data/config.json format, and first commit format that are easy to get wrong without this reference. Invoke whenever the user is starting from a fresh or empty repository, has never built a Keboola component before, wants to run the cookiecutter template, or says things like "new extractor for X", "build ex-something from scratch", "fresh repo", "empty git repo", "brand new component", "never done this in keboola", "initialize a component", "scaffolding a new component", "just initialized the git repo", "nothing set up yet". Not for existing components that already have src/component.py.---# Initialize New Keboola ComponentBefore writing any code, check whether the component has already been initialized.## Step 1: Detect stateCheck for signs of an existing cookiecutter setup:```bashls src/component.py component_config/ pyproject.toml 2>/dev/null```- If all present → skip initialization, hand off to `@develop-component` immediately- If missing or empty repo → proceed with initialization below## Step 2: Gather component infoYou may already know the component details from the conversation. If not, ask:- **Component name** — kebab-case, no type suffix (e.g. `ex-salesforce`, not `salesforce-extractor`)- **Component ID** — `vendor.component-name` format (e.g. `keboola.ex-salesforce`)- **Type** — `extractor`, `writer`, or `application`- **Short description** — one sentence of what it doesNaming rules:- Never include "extractor", "writer", or "application" in the name itself- Use prefix convention: `ex-` for extractors, `wr-` for writers## Step 3: Run cookiecutter```bash# Install if missingwhich cookiecutter || pip install --user cookiecuttercookiecutter gh:keboola/cookiecutter-python-component```Run interactively so the user can confirm the values. The template creates `src/`, `component_config/`, `tests/`, `.github/workflows/`, `Dockerfile`, `pyproject.toml`, and `data/` with example files.## Step 4: Clean up and configure```bash# Remove the generic example files that come with the templatefind data -type f -delete# Recreate the expected directory structuremkdir -p data/in/tables data/in/files data/out/tables data/out/files```Then create `data/config.json` with realistic example parameters based on what the component actually does — not a generic placeholder. Read `component_config/configSchema.json` if it already has content, and mirror the required fields with plausible example values. Use `#` prefix for sensitive fields (e.g. `"#api_key"`).## Step 5: Initial commit```bashgit add -Agit commit -m "feat: initialize component from cookiecutter template"```## Step 6: Hand offOnce initialized, pass to `@develop-component` to start the implementation. Share what you know about the component's purpose so it has context.---## Key resources- Cookiecutter template: `gh:keboola/cookiecutter-python-component`- Developer Portal: https://components.keboola.com/- Component tutorial: https://developers.keboola.com/extend/component/tutorial/

Frequently asked questions