Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/LIDR-academy/lidr-specboot/llms.txt

Use this file to discover all available pages before exploring further.

Specboot gives your project a production-ready AI development foundation in seconds. Run one npx command and your project gains consistent standards for Claude, Cursor, GitHub Copilot, and Gemini — all pointing to a single source of truth — plus a library of reusable agent skills and a battle-tested spec-driven workflow.

Quickstart

Bootstrap Specboot into your project and run your first AI workflow in minutes.

Core Concepts

Understand spec-driven development and how Specboot’s architecture works.

Workflow Guide

Walk through the full OpenSpec command flow from user story to merged PR.

Skills Reference

Explore the library of reusable AI agent skills included with Specboot.

What is Specboot?

Specboot is a portable CLI tool and template kit that bootstraps an augmented spec-driven development environment into any project. At its core it solves a common problem: when a team uses multiple AI coding assistants, each tool operates in isolation with different (or no) context about the project’s standards, architecture, and workflows. Specboot fixes this by:
  • Installing a single source of truth (docs/base-standards.md) that all AI copilots read automatically
  • Wiring up symlinks so Claude, Cursor, Codex, and Gemini each find their configuration at their expected path
  • Providing pre-built agent roles (backend developer, frontend developer, product analyst) with expert-level system prompts
  • Bundling reusable skills for recurring tasks like enriching user stories, auditing code, creating commits, and managing git worktrees
  • Integrating with OpenSpec for a structured spec-driven loop: plan → specify → implement → verify → ship
1

Install Specboot

Run npx @lidr/lidr-specboot in your project root. Files are copied and symlinks are created automatically — existing files are never overwritten.
2

Customize your docs/

Update the files in docs/ with your project’s real stack, architecture patterns, API contracts, and domain models. This step is mandatory for meaningful AI assistance.
3

Initialize OpenSpec

Run openspec init and point your config.yml to the docs/ and ai-specs/ directories Specboot just installed.
4

Start your first workflow

Use /enrich-us to refine a user story, /ff to generate feature artifacts, /apply to implement tasks, and /verify to validate the result.

Key Features

Multi-Copilot Support

A single docs/base-standards.md is symlinked to CLAUDE.md, AGENTS.md, codex.md, and GEMINI.md so every AI tool loads the same rules.

AI Agent Roles

Pre-built agent definitions for backend (DDD/TypeScript), frontend (React), and product strategy work.

Reusable Skills

Composable skill prompts for enriching stories, auditing code, writing commits, managing worktrees, and more.

Development Standards

Opinionated TDD-first standards with 90%+ coverage requirements, strict TypeScript, and clean architecture patterns.

Build docs developers (and LLMs) love