project-init

Scan repositories to generate project-specific CLAUDE.md files for AI agents.

7|1|Updated Feb 12, 2026
One-click install
npx skills add https://github.com/chdalski/claude_orchestration --skill project-init-chdalski
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: project-init
Source: https://github.com/chdalski/claude_orchestration/tree/main/blueprint_v2/.claude/skills/project-init
Command: npx skills add https://github.com/chdalski/claude_orchestration --skill project-init-chdalski

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill removes the manual, error-prone work of creating and maintaining project context files for Claude Code agents by scanning a repository and producing accurate CLAUDE.md documents so agents boot with precise, project-specific grounding.

Core Features & Use Cases

  • Automatic language and framework detection from manifest files at the project root and one level of subdirectories.
  • Mono-repo and sub-project handling that catalogs workspace members, detects git submodules, and generates per-subproject CLAUDE.md when needed.
  • Rule mapping and lint application, including reading language-specific init files (e.g., rust-init.md) to apply Cargo lints and merging existing lint sections safely.
  • Safe write behavior and reporting that summarizes detected languages, active rule files, TODO sections that need human input, and any modified manifests so leads can decide follow-up tasks.
  • Use Case: Onboard a new codebase into a Claude orchestration blueprint, or regenerate agent context after major structural changes.

Quick Start

Run the project-init skill to scan the repository and generate CLAUDE.md files with detected languages, frameworks, active rules, and suggested lint updates.

Frequently Asked Questions about project-init

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

FAQPage Schema
How do I generate a CLAUDE.md file for my project automatically?

To generate a CLAUDE.md file automatically, you can use a repository scanning tool that reads manifest files to detect languages and frameworks, mapping them to conditional rule files to produce accurate project context for AI agents.

How does language detection work when generating project context for AI agents?

Language detection works by reading manifest files located at the project root and one subdirectory level deep. This process identifies the specific languages and frameworks in use to map them to appropriate conditional rule files.

Can I generate CLAUDE.md files for a mono-repo with git submodules?

Yes, you can generate context files for a mono-repo with git submodules. The scanning process catalogs workspace members, detects nested submodules, and produces per-subproject CLAUDE.md files as needed.

How do I apply Cargo lints when generating a Rust project context?

To apply Cargo lints for a Rust project, the generation process reads language-specific init files like rust-init.md, applying the lints and safely merging them into existing lint sections within the context file.

Will generating a CLAUDE.md file overwrite my existing project context?

No, generating a context file will not overwrite an existing CLAUDE.md without explicit user confirmation. The process includes safe write behavior and reports TODO sections requiring human input.

When should I regenerate project context files for an AI agent?

You should regenerate project context files when onboarding a new codebase into an AI orchestration blueprint or after major structural changes to the repository, ensuring the agent retains accurate grounding.