init

Create Accelerator meta directories and configure .gitignore entries in repositories.

30|1|Updated Mar 15, 2026
One-click install
npx skills add https://github.com/atomicinnovation/accelerator --skill init-atomicinnovation
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: init
Source: https://github.com/atomicinnovation/accelerator/tree/main/skills/config/init
Command: npx skills add https://github.com/atomicinnovation/accelerator --skill init-atomicinnovation

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prepares a repository with the directory structure and .gitignore entries that Accelerator skills expect so the plugin can read and write shared artifacts reliably. It eliminates manual setup and the risk of missing meta folders or incorrect ignore rules that would break the Accelerator workflow.

Core Features & Use Cases

  • Idempotent repository bootstrapping: Creates required meta directories and .gitkeep files while reporting which items were created versus already present.
  • Safe tmp configuration: Ensures the tmp directory contains an internal .gitignore that preserves the directory while ignoring its contents.
  • Gitignore management: Adds .claude/accelerator.local.md to the repository root .gitignore if missing, preserving existing file formatting.
  • Use Case: Run once when adopting Accelerator in a project or re-run when onboarding a new contributor to ensure consistent filesystem-based shared memory.

Quick Start

Run the init skill to create Accelerator meta directories, set up tmp ignore behavior, and ensure the .claude/accelerator.local.md entry is present in the repository root .gitignore.

Frequently Asked Questions about init

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

FAQPage Schema
How do I bootstrap repository directories and gitignore for a new project?

To bootstrap repository directories and gitignore, this skill creates required meta folders, adds .gitkeep placeholders, and configures tmp .gitignore rules. It safely writes or appends files without modifying unrelated content in your root .gitignore.

What is the best way to set up shared meta directories in an existing code repository?

Setting up shared meta directories in an existing repository requires generating specific folder structures and ignore rules. This skill resolves target paths via plugin configuration and performs idempotent existence checks to create missing directories without altering existing project files.

Why does my gitignore not preserve the tmp directory while ignoring its contents?

Preserving a tmp directory while ignoring its contents requires an internal .gitignore file with specific ignore patterns. This skill safely configures the tmp directory by writing an internal .gitignore that keeps the folder tracked while excluding its generated artifacts.

Can I use an init script to add local accelerator entries to my root gitignore?

Yes, you can use an init script to add local accelerator entries to your root .gitignore. This skill appends the .claude/accelerator.local.md entry to the root .gitignore if it is missing, preserving the existing file formatting and unrelated rules.

Does running repository initialization multiple times duplicate gitignore entries or overwrite files?

Running repository initialization multiple times does not duplicate gitignore entries or overwrite files. The skill performs idempotent existence checks on every target path, reporting which items were created versus already present, ensuring safe repeated execution.

When do I need to run repository bootstrapping for Accelerator skills?

You need to run repository bootstrapping when adopting Accelerator in a project or onboarding a new contributor. This ensures consistent filesystem-based shared memory by creating the meta directories and gitignore entries that Accelerator skills expect to read and write reliably.