project-init

Generate agent-fix config.yaml by detecting project structure, tooling, and auth inputs.

1|Updated Apr 9, 2026
One-click install
npx skills add https://github.com/noni2she/agent-fix --skill project-init-noni2she
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: project-init
Source: https://github.com/noni2she/agent-fix/tree/main/skills/project-init
Command: npx skills add https://github.com/noni2she/agent-fix --skill project-init-noni2she

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the manual setup work required to prepare an agent-fix run by automatically detecting a target project's structure, quality commands, dev server details, and (when applicable) frontend auth configuration.

Core Features & Use Cases

  • Project structure detection: Scans project_path, derives a kebab-case slug from package.json, and identifies monorepo vs single-app layouts.
  • Context file inventory: Registers available context documents (e.g., CLAUDE.md, AGENTS.md, spec files, API specs, and README.md) into context_sources.md.
  • Config.yaml generation: Produces projects/<slug>/config.yaml including framework, language, quality_checks, issue_source, skills directories, behavior_validation, and MCP server settings.
  • Quality tooling discovery: Detects type-check/typecheck, lint/eslint, prettier, and test commands to populate verification steps.
  • Frontend auth wiring (best-effort): If the project looks like a frontend framework, searches for login/auth routes and modal triggers, and fills selector fields; otherwise it disables behavior validation and auth details.

Quick Start

Ask the agent to run project-init against your repository path and generate projects/<slug>/config.yaml plus context_sources.md.

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 config.yaml for agent-fix runs in a monorepo?

Config.yaml generation scans your project_path to detect monorepo or single-app layouts, deriving a kebab-case slug from package.json to output a configuration file under projects/<slug>/.

How does auto-detection infer framework and quality commands from existing tooling?

Framework inference and quality tooling discovery scan context source files to identify type-check, lint, prettier, and test commands, populating these verification steps directly into your config.yaml.

Can I automatically wire frontend auth selectors for Playwright behavior validation?

Yes, frontend auth wiring searches for login routes and modal triggers to fill Playwright auth selectors, but behavior validation is automatically disabled if frontend auth evidence cannot be reliably detected.

Does MCP integration work with both single-app and monorepo project structures?

MCP integration works with both monorepo and single-app layouts by scanning the target project structure and registering MCP server settings within the generated config.yaml output.

What context files are registered during project setup configuration generation?

Context file inventory registers available documents such as CLAUDE.md, AGENTS.md, spec files, API specs, and README.md into a context_sources.md file during project setup configuration generation.

When should I not use automated config generation for issue reproduction?

You should not rely on automated config generation when frontend-only behavior_validation is required but auth or browser evidence cannot be reliably detected, as the Skill will automatically disable these fields.