project-init

Automate new research project setup with conda environment and standardized directory structure.

2|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/Gonglitian/agent-skills --skill project-init-gonglitian
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: project-init
Source: https://github.com/Gonglitian/agent-skills/tree/main/skills/project-init
Command: npx skills add https://github.com/Gonglitian/agent-skills --skill project-init-gonglitian

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Initialize a new research project with all standard infrastructure: a reproducible environment, a CLAUDE.md AI guide, git setup, and a clean folder structure with organized data and checkpoints.

Core Features & Use Cases

  • Create a conda environment and basic project scaffold for data science and ML research.
  • Generate CLAUDE.md with suggested environment details, data paths, and common commands.
  • Establish a reusable directory layout (data, checkpoints, logs, docs, scripts, papers, outputs) and symlink to data disks.
  • Configure Git workflow, ignore rules, and initial planning/docs for project planning.
  • Trigger automatically on project start, clone, or explicit commands like 新项目, 初始化, project init.

Quick Start

Start a new project by invoking the skill with a trigger phrase such as 新项目 or 初始化 to set up a new project.

Frequently Asked Questions about project-init

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

FAQPage Schema
How do I initialize a research project with a standard directory layout and conda environment?

To initialize a research project, this skill automates creating a per-project conda environment alongside a standard directory layout containing data, checkpoints, and logs, ensuring your workspace is ready for immediate use.

How do I set up a CLAUDE.md file and initial planning docs for a new machine learning project?

Setting up a CLAUDE.md file is handled automatically during project initialization, generating an AI guide with suggested environment details, data paths, and common commands alongside initial planning documents for your workflow.

Does initializing a research project with this method configure git workflow and ignore rules?

Yes, initializing a research project with this method configures the git workflow, establishes ignore rules, and sets up initial documentation to maintain a clean and reproducible repository structure.

What is the best way to symlink data disks into a standardized research project scaffold?

The best way to symlink data disks is through automated project scaffolding, which creates symlinks to your data disks within a reusable directory layout for data, checkpoints, logs, and outputs.

Can I trigger project initialization automatically when cloning an existing repository?

Yes, you can trigger project initialization automatically when cloning an existing repository or starting a new project, applying standardized scaffolding to enforce environments and directory structures.