fresh-start

Detect project root, mode, and validate files for execution context.

39|5|Updated Dec 28, 2025
One-click install
npx skills add https://github.com/benjaminshoemaker/ai_coding_project_base --skill fresh-start
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: fresh-start
Source: https://github.com/benjaminshoemaker/ai_coding_project_base/tree/main/.claude/skills/fresh-start
Command: npx skills add https://github.com/benjaminshoemaker/ai_coding_project_base --skill fresh-start

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you quickly initialize a project workspace by detecting the project root, determining the workflow mode (greenfield or feature), loading the appropriate context, and verifying essential scaffolding before execution. It reduces setup friction and ensures consistent state across phases.

Core Features & Use Cases

  • Context detection: infer PROJECT_ROOT, WORKING_DIR, and MODE (greenfield or feature) from your current folder or an optional path.
  • Readiness validation: verify essential files (AGENTS.md, EXECUTION_PLAN.md) exist and guide users to fix gaps.
  • Git and branch awareness: identify the current branch, summarize recent commits, and flag uncommitted changes to keep work aligned.
  • Phase-state handling: detect existing phase-state.json to auto-resume progress and pick up where you left off.

Quick Start

Run /fresh-start in your project directory to initialize the context and load the current project state. If you are in a feature folder, the tool will adapt to feature mode and use the corresponding feature path. After running this, you can proceed to generate plans and executions.

Frequently Asked Questions about fresh-start

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

FAQPage Schema
How do I initialize project context and detect the workspace root for a greenfield project?

To initialize project context, run the command in your directory to automatically detect the project root, infer the working directory, and set the mode to greenfield. It validates required scaffolding files and loads the current state for execution.

What is the best way to load execution context and check git status before starting feature work?

Loading execution context is handled by orienting to your project directory, which automatically determines the mode, identifies the current git branch, summarizes recent commits, and flags uncommitted changes to align your work.

Does the project setup workflow require AGENTS.md and EXECUTION_PLAN.md files?

Yes, the project setup workflow requires AGENTS.md and EXECUTION_PLAN.md files. The readiness validation step checks for these essential files and guides you to fix any gaps before proceeding with execution.

Can I resume an interrupted workflow by detecting the existing phase state?

Yes, you can resume an interrupted workflow by detecting an existing phase-state.json file. The tool checks for phase state to automatically resume progress and pick up exactly where you left off.

How do I adapt context loading for a project directory containing feature folders?

Context loading adapts automatically by detecting if you are in a feature folder. It switches to feature mode and uses the corresponding feature path, ensuring the correct context is loaded for that specific workflow.