repo-scaffold

Scaffolds runnable repositories with agent-facing documentation from prompts, conversations, or documents.

Updated Dec 7, 2025
One-click install
npx skills add https://github.com/harlanljones/dotfiles --skill repo-scaffold-harlanljones
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: repo-scaffold
Source: https://github.com/harlanljones/dotfiles/tree/main/dot_hermes/skills/repo-scaffold
Command: npx skills add https://github.com/harlanljones/dotfiles --skill repo-scaffold-harlanljones

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Starting a new project from a loose idea, chat transcript, or spec often produces inconsistent boilerplate and documentation that a fresh agent or developer cannot act on without re-deriving context. This Skill turns any input into a working, tested repository skeleton whose docs are written for agents first. ## Core Features & Use Cases - Brief-driven scaffolding: Extracts purpose, stack, constraints, and non-goals from prompts, prior conversations, and documents before writing any files. - Minimal-but-runnable skeleton: Creates dependency manifests, an entry point, a passing test, lint config, and an initial git commit, then verifies install, test, and run commands actually work. - Agent-facing documentation: Generates AGENTS.md, ARCHITECTURE.md, CONVENTIONS.md, and ROADMAP.md with verified copy-paste commands and a layout map. - Use Case: Paste a product spec and ask for a project; receive a git-initialized repo where tests pass and AGENTS.md lets a cold-start agent build, test, and extend the code immediately. ## Quick Start Turn this specification document into a fully scaffolded repository with agent-facing documentation and a verified runnable skeleton.

Frequently Asked Questions about repo-scaffold

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

FAQPage Schema
How do I scaffold a new repository from a specification document?▼

Provide the spec, prompt, or conversation as input; the skill extracts a brief covering purpose, stack, constraints, and non-goals, then plans the full file tree before creating anything. It scaffolds a minimal runnable skeleton with tests, lint config, and an initial git commit.

What is AGENTS.md and why write documentation for AI agents?▼

AGENTS.md is a repo-root operating manual for AI agents covering environment setup, verified commands, layout map, conventions, and non-goals. It lets a fresh agent session act on the codebase without re-deriving context from the original conversation.

When should I not use repo scaffolding?▼

Avoid it when adding a module to an existing repository, where you should follow that repo's own conventions instead. It is also unsuited for single-file scripts, which are better handled by a lightweight spike or prototype approach.

How does the scaffold verify the project actually works?▼

The skill runs install, test, and entry-point commands and records their output before finishing. It also re-reads AGENTS.md cold and executes every command it references, and cross-checks that documented files exist and existing files are documented.

Can scaffolding handle ambiguous requirements like language or framework choice?▼

If an ambiguous decision changes the file layout, such as language, framework, or monorepo versus single package, the skill asks the user once for clarification. Otherwise it picks the obvious default and records the assumption in the brief.