guardrails

Generate domain-specific guardrail rules into .claude/rules/*.md during bootstrap.

3|Updated Mar 16, 2026
One-click install
npx skills add https://github.com/cardinalconseils/claude-starter --skill guardrails-cardinalconseils
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: guardrails
Source: https://github.com/cardinalconseils/claude-starter/tree/main/skills/guardrails
Command: npx skills add https://github.com/cardinalconseils/claude-starter --skill guardrails-cardinalconseils

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Guardrails needed to enforce security and best practices are often manually created during bootstrap; this skill automates the generation of domain guardrail rules to codify security, testing, database, and documentation standards.

Core Features & Use Cases

  • Generates .claude/rules/{security,testing,database,docs}.md guardrail files tailored to detected project concerns.
  • Replaces placeholders with detected values (AUTH_METHOD, API_DIRECTORY, TEST_FRAMEWORK, DB_CLIENT) and ensures valid globs.
  • Catalog-driven customization to provide concise, actionable guardrails aligned to the project's stack.

Quick Start

Run bootstrap and rely on the guardrails skill to produce initial guardrail files from your detected stack.

Frequently Asked Questions about guardrails

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

FAQPage Schema
How do I auto-generate security and code-quality guardrails during project bootstrap?

To auto-generate security and code-quality guardrails during bootstrap, run a bootstrap scan to detect project values, then use catalog-driven rules to write tailored .claude/rules/*.md files with valid globs and replaced placeholders.

What are bootstrap guardrail rules and when do I need them for my project?

Bootstrap guardrail rules are .claude/rules/*.md files that codify security, testing, database, and docs standards. You need them during project initialization to enforce domain-specific best practices automatically from detected stack values.

Can I customize guardrail files for specific testing frameworks and database clients?

Yes, you can customize guardrail files by loading catalogs that replace placeholders like TEST_FRAMEWORK, DB_CLIENT, API_DIRECTORY, and AUTH_METHOD with values detected from your scan results, ensuring concise and actionable rules.

How does the guardrails skill validate frontmatter structure and globs in generated rules?

The guardrails skill validates frontmatter structure and ensures valid globs by scanning your project during bootstrap, loading rule catalogs, and writing .claude/rules/*.md files with placeholders correctly replaced by detected domain values.

Do I need external dependencies or components to generate domain guardrail rules?

No external dependencies or components are required to generate domain guardrail rules. The skill operates independently during bootstrap, using scan results to produce security, testing, database, and documentation .md rule files.

What's the best way to enforce documentation and database standards at project initialization?

The best way to enforce documentation and database standards at initialization is automating guardrail generation, which writes tailored .claude/rules/{database,docs}.md files from detected stack values to codify project standards without manual setup.