add

Orchestrate AI-driven development workflows with state-tracked task progression.

2|Updated Mar 26, 2026
One-click install
npx skills add https://github.com/pilotspace/moon --skill add-pilotspace
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: add
Source: https://github.com/pilotspace/moon/tree/main/.claude/skills/add
Command: npx skills add https://github.com/pilotspace/moon --skill add-pilotspace

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

ADD provides a disciplined orchestration for AI-driven development, ensuring direction, safety, and auditable progress from intake to observe.

Core Features & Use Cases

  • One-file-per-task orchestration: each feature lives under .add/tasks/<slug>/TASK.md with seven sections.
  • Brownfield-friendly onboarding and baseline approval to start a build safely.
  • Supports multi-phase flow (specify, scenarios, contract, tests, build, verify, observe) with evidence-driven gates.
  • On-demand resources and living docs to keep teams aligned and auditable.

Quick Start

Initialize ADD in your repository and then confirm the baseline approval to lock the foundation and start the workflow.

Frequently Asked Questions about add

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

FAQPage Schema
How do I orchestrate AI-driven development workflows safely in a repository?

Orchestrate AI-driven development safely by initializing a .add directory to manage tasks, contracts, tests, and builds. This enforces structured, state-tracked progression from specification to observation with evidence-based verification gates.

What is human baseline approval in AI DevOps automation?

Human baseline approval is a mandatory safety rule that locks the foundation before execution. It freezes the contract after brownfield onboarding, ensuring AI builds only proceed with per-scope autonomy after explicit human confirmation.

Can I use structured task orchestration for existing brownfield projects?

Yes, this workflow supports brownfield-friendly onboarding. You can initialize the orchestration structure in existing repositories, requiring a baseline approval to safely lock the foundation before starting any AI-driven build phases.

How do I track AI development progress from specification to observation?

Track AI development progress using one-file-per-task orchestration under .add/tasks with seven sections. The multi-phase flow moves through specify, scenarios, contract, tests, build, verify, and observe with evidence-driven gates.

What are the limitations of state-tracked AI workflow automation?

State-tracked AI workflow automation requires a .add directory and strict adherence to evidence-driven gates. Autonomy is constrained per-scope, and execution cannot proceed without freezing contracts and securing prior human baseline approval.