setup-init

Initializes the plugin in a project by running bootstrap scripts and orchestrating LLM setup steps.

1|Updated Jun 27, 2026
One-click install
npx skills add https://github.com/vickysrawat/AI-Assisted-development --skill setup-init-vickysrawat
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: setup-init
Source: https://github.com/vickysrawat/AI-Assisted-development/tree/main/skills/setup-init
Command: npx skills add https://github.com/vickysrawat/AI-Assisted-development --skill setup-init-vickysrawat

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Setting up the ai-assisted-development plugin in a new project involves many mechanical steps (directories, hooks, command stubs, gitignore, state files) plus LLM-driven work (CLAUDE.md generation, architecture analysis, knowledge graph). This Skill automates the entire initialization in one idempotent, resumable workflow. ## Core Features & Use Cases - Deterministic bootstrap: Runs a single script that creates directories, 37 command stubs, enforcement hooks, state files, and gitignore entries — crash-safe and safe to re-run. - Repo-type detection: Detects the project type (.NET, Angular, React, Spring Boot, Python, etc.) before any work begins, failing fast on unrecognizable repositories. - Orchestrated LLM steps: Choreographs /init, the architect skill, CLAUDE.md cleanup routing, and knowledge-graph creation, tracking progress in a resume manifest. - Use Case: A developer clones a .NET + Angular monorepo and runs /setup-init; the skill provisions all plugin infrastructure, generates architecture docs, and builds the knowledge graph without manual configuration. ## Quick Start Run /setup-init in a new project to bootstrap the plugin, generate CLAUDE.md and architecture docs, and build the knowledge graph.

Frequently Asked Questions about setup-init

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

FAQPage Schema
How do I set up the ai-assisted-development plugin in a new project?

Run /setup-init once in the project root. It detects the repo type, runs the bootstrap script for all mechanical work, then orchestrates /init, the architect skill, and knowledge-graph creation automatically.

Is it safe to re-run /setup-init after a failure?

Yes, setup-init is fully idempotent and resume-safe. A bootstrap manifest tracks completed work, so re-running skips finished steps and resumes pending LLM tasks without overwriting existing files.

What project types does setup-init support?

It detects dotnet-api, aspnet-framework, aspnet-mvc, angular-nx, angular-standard, react, js-library, spring-boot, vsto, python-fastapi, python-django, and python-flask. Ambiguous repos trigger an LLM fallback; unrecognizable repos hard-fail with a clear error.

What happens if repo type detection fails during setup?

The skill hard-fails before any bootstrap work with a message explaining no recognizable source files were found. You can tell Claude the project type explicitly and re-run, or add source files first.

Can I skip the enforcement hooks during setup-init?

Yes, pass --no-hooks to the bootstrap command to skip hooks, PreToolUse wiring, and the pre-commit hook. The opt-out must be recorded in .claude/architecture/architecture-deployment.md per ADR 0009.