fdl-worktree-create

Create isolated farming-data-layer git worktrees with per-branch dbt-target state.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/janezlapajne/dotfiles --skill fdl-worktree-create
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: fdl-worktree-create
Source: https://github.com/janezlapajne/dotfiles/tree/main/conf/dotfiles/claude/skills/fdl-worktree-create
Command: npx skills add https://github.com/janezlapajne/dotfiles --skill fdl-worktree-create

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents dev collisions and messy histories by creating an isolated worktree for each farming-data-layer change, with a dedicated dbt-target state per branch.

Core Features & Use Cases

  • Isolated per-task worktrees: Creates a dedicated worktree so parallel investigations and fixes don’t overwrite each other’s dbt-target state in BigQuery.
  • MR-aware branching workflow: Reuses an existing linked MR only if its branch is effectively empty; otherwise it creates a new Draft MR while never creating a new issue.
  • Strict safety guardrails: Refuses to run from inside an existing non-main worktree and requires upfront preparation (sets dbt-target before switching).

Quick Start

Tell the AI to run fdl-worktree-create when you paste a GitLab issue or MR URL and you want an investigation worktree started (for example: “Create a worktree for this issue and prepare my dbt-target.”).

Frequently Asked Questions about fdl-worktree-create

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

FAQPage Schema
How do I set up an isolated git worktree for a new dbt feature branch?

To set up an isolated git worktree, you provide a GitLab issue or MR URL and the Skill creates a dedicated branch, links a Draft MR, and assigns a per-branch dbt-target state var. This prevents parallel investigations from overwriting each other's BigQuery target state.

Can I reuse an existing GitLab MR branch when creating a worktree?

You can reuse an existing linked MR branch only if it has zero commits ahead of origin/main. If the branch already contains changes, the workflow requires creating a new Draft MR instead to maintain a clean history.

What are the limitations of running worktree creation inside an existing worktree?

A strict safety limitation is that the workflow refuses to run from inside an existing non-main worktree. You must execute the setup process from your main checkout to prevent nested branching conflicts and state corruption.

How do I prepare a BigQuery dbt-target before switching into a new worktree?

The workflow enforces upfront preparation by setting the per-branch dbt-target state var before switching into the created worktree. This ensures your BigQuery target configuration is fully mapped before any development begins.