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 installs as a project-level bootstrap, not as a permanent global dependency. Run it once inside your project and it copies all template files, wires up the multi-copilot symlinks, and exits — leaving behind a self-contained directory structure you own and version-control alongside your code.

​
Requirements

  • Node.js >=18 — required by the @lidr/lidr-specboot package (engines.node field in package.json). Node.js 20.19.0 or higher is recommended if you also plan to use OpenSpec.
  • npm npx — included with all npm installations; no separate install needed.
  • An existing project directory. Specboot copies files into it; it does not scaffold a new project from scratch.

​
Installation Methods

# Run from inside your project directory
npx @lidr/lidr-specboot
Method 1 — npx (recommended): Fetches the latest published package from npm, copies the full template into the current working directory, and automatically creates all symlinks. No global install required. Safe to run more than once. Method 2 — npx with a target directory: Identical to Method 1, but you specify the destination path as an argument instead of cd-ing into it first. Useful in scripts or CI bootstrapping. Method 3 — Global install: Installs the lidr-specboot binary globally via npm install -g @lidr/lidr-specboot, then run lidr-specboot (with an optional target directory argument) from anywhere. Useful when you bootstrap multiple projects regularly and prefer not to pull from the npm registry each time. Method 4 — Manual copy: Use this when you need offline installation or want to inspect the files before they land in your project. The -n flag in cp prevents overwriting any files that already exist. After copying, create the four copilot symlinks by hand as shown above, then mirror ai-specs/agents/ and ai-specs/skills/ into .claude/ and .cursor/ using relative symlinks (see Symlink Structure below).

​
What Gets Installed

After a successful run, your project will contain: 
your-project/
├── docs/
│   ├── base-standards.md           # Core dev rules — single source of truth
│   ├── backend-standards.md        # API, database, security, testing
│   ├── frontend-standards.md       # React, UI/UX, state management
│   ├── documentation-standards.md  # Doc structure, formatting, AI standards
│   ├── api-spec.yml                # OpenAPI 3.0 reference spec
│   ├── data-model.md               # Database schemas and domain models
│   └── development_guide.md        # Setup and workflow guide
├── ai-specs/
│   ├── agents/                     # Agent role definitions
│   │   ├── backend-developer.md
│   │   ├── frontend-developer.md
│   │   └── ...
│   └── skills/                     # Reusable skill workflows (SKILL.md files)
│       ├── enrich-us/
│       ├── using-git-worktrees/
│       ├── code-auditing/
│       └── ...
├── AGENTS.md  -> docs/base-standards.md
├── CLAUDE.md  -> docs/base-standards.md
├── codex.md   -> docs/base-standards.md
└── GEMINI.md  -> docs/base-standards.md
Specboot creates two layers of symbolic links so every supported copilot can discover agents and skills using the paths it already expects. Layer 1 — Copilot configuration files (project root) The four root-level configuration files are symlinks pointing to docs/base-standards.md: 
CLAUDE.md  -> docs/base-standards.md
AGENTS.md  -> docs/base-standards.md
codex.md   -> docs/base-standards.md
GEMINI.md  -> docs/base-standards.md
All copilots read the same rules from one file. Update base-standards.md once and every tool benefits. Layer 2 — Agent and skill discovery (.claude/ and .cursor/) Each agent and skill in ai-specs/ is mirrored into .claude/ and .cursor/ via relative symlinks so tool-specific UIs can discover them without duplicating files: 
.claude/agents/<agent>  -> ../../ai-specs/agents/<agent>
.claude/skills/<skill>  -> ../../ai-specs/skills/<skill>
.cursor/agents/<agent>  -> ../../ai-specs/agents/<agent>
.cursor/skills/<skill>  -> ../../ai-specs/skills/<skill>
ai-specs/ is always the canonical source. Never edit the files inside .claude/ or .cursor/ directly — edit the originals in ai-specs/ and the symlinks reflect changes automatically.

​
Safe Re-Run Behavior

Specboot is idempotent. If you run npx @lidr/lidr-specboot a second time in a project that already has Specboot installed, every file and symlink that already exists is skipped — nothing is overwritten, nothing is deleted. The summary output shows a Skipped count so you can confirm no changes were made unexpectedly. This makes it safe to run in CI or to use as a verification step after a merge.

​
Verify the Installation

After running Specboot, confirm the symlinks are correctly in place: 
# Check the four root-level copilot symlinks
ls -la CLAUDE.md AGENTS.md codex.md GEMINI.md
Expected output: 
AGENTS.md -> docs/base-standards.md
CLAUDE.md -> docs/base-standards.md
GEMINI.md -> docs/base-standards.md
codex.md  -> docs/base-standards.md

# Verify the .claude skill and agent symlinks
ls -la .claude/agents/
ls -la .claude/skills/

# Confirm docs/ and ai-specs/ directories were created
ls docs/
ls ai-specs/agents/
ls ai-specs/skills/
If any symlink points to a missing target (shown as a broken link in ls -la), re-run npx @lidr/lidr-specboot — it will create any missing links while skipping everything that already exists.
Customizing docs/ is mandatory. The files installed by Specboot are generic reference examples from the LIDR project. Before using Specboot with any AI copilot, you must update docs/base-standards.md and the companion standards files to reflect your project’s real tech stack, architecture patterns, domain language, API contracts, and data models. Skipping this step means every AI agent operates on incorrect context and will produce code that does not match your project. See the Customization Guide for step-by-step instructions and ready-to-use prompts.

Build docs developers (and LLMs) love

Get started for freeTalk to us