Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    AI Coding Guardrails:Ultimate .curcorrules, Claude generator

    1

    Generate evidence-based coding instructions and scoped rules for Cursor and Claude Code based on your repository.

    $8

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    AI Coding Guardrails:Ultimate .curcorrules, Claude generator

    AI Coding Guardrails:Ultimate .curcorrules, Claude generator

    Example session with this skill installed

    Audit my repo and generate a CLAUDE.md file and a Cursor rule for my TypeScript React components in src/components.

    • Read your context and instructions
    • Compiled the ai coding guardrails

    I have inspected the repository. I found a Vite-based React setup using pnpm and vitest. I am generating CLAUDE.md for general project context and a scoped Cursor rule for your components.

    AI Guardrails Generated

    
# CLAUDE.md
## Project Overview
- Tech Stack: React, TypeScript, Vite, Tailwind CSS
- Package Manager: pnpm

## Build & Test Commands
- Build: `pnpm build`
- Dev: `pnpm dev`
- Test: `pnpm test`
- Lint: `pnpm lint`

## Coding Conventions
- Use functional components with TypeScript interfaces for props
- Follow the atomic design pattern in `src/components`
- Use Tailwind for styling; avoid CSS modules unless specified

    ---
description: Rules for React components in src/components
globs: src/components/**/*.tsx
alwaysApply: false
---

# React Component Standards

## Patterns
- Use `interface Props` for component properties.
- Export components as named exports.
- Place associated types in the same file as the component.

## Verification
- Run `pnpm test <file>` after modifying component logic.
- Run `pnpm lint` to verify styling and type constraints.

    Next steps

    1. Review the generated CLAUDE.md in the project root.
    2. Move the .mdc content into .cursor/rules/react-components.mdc.
    3. Run pnpm test to ensure the listed test commands are fully operational.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Writing custom instructions for AI coding assistants by hand is tedious and prone to error. This skill inspects your actual repository structure, scripts, and conventions to generate precise, evidence-based guardrails for Cursor and Claude Code.

    What it does

    • Evidence-based auditing reads your package manifests, README, and source files to extract real project conventions.
    • Multi-target generation creates .cursor/rules/*.mdc files for modern Cursor and CLAUDE.md for Claude Code.
    • Legacy migration handles the conversion from root .cursorrules to scoped .mdc rules with proper YAML frontmatter.
    • Context optimization structures rules using globs and descriptions to ensure the LLM only sees instructions relevant to the current file.
    • Command verification validates every suggested terminal command against your actual scripts or binaries before inclusion.

    How it works

    1. Repository Inspection starts by scanning your build configs, test suites, and project architecture.
    2. Drafting creates scoped instruction files based on the verified tech stack and directory patterns.
    3. Refinement merges new rules with existing ones, preserving your manual edits while fixing contradictions.
    4. Validation checks YAML frontmatter, file globs, and cross-references for syntax and logic errors.

    Frameworks & tools

    Works with any repository architecture. Specifically optimized for Cursor (.mdc, .cursorrules) and Claude Code (CLAUDE.md, .claude/rules/).

    Why this beats prompting it yourself

    Generic prompts produce boilerplate "best practices" that often conflict with your actual build scripts or file paths. This skill generates instructions grounded in your specific codebase, reducing LLM hallucinations and ensuring the agent uses your preferred package manager and test runners correctly.

    Use cases

    • Scaffold a new CLAUDE.md and .cursor/rules/ directory for a fresh project.
    • Migrate a bloated legacy .cursorrules file into organized, scoped .mdc rules.
    • Audit existing instructions to remove broken links or commands that no longer exist in the repo.
    • Standardize AI coding conventions across a monorepo with specific directory-level rules.

    Known limitations

    Does not create automated hooks, CI/CD scripts, or tool permissions. Instructions are context for the model, not hard technical enforcement or security boundaries.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Prompt & Skill Engineering.

    Instructions provide context, not technical enforcement. Cannot create automated scripts or modify tool permissions. Requires read access to project manifests and source files.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 22 days ago

    Permissions

    Read Files
    Write Files
    Listed22 days ago

    What's inside

    Frequently Asked Questions

    Unlike manual prompting, this skill audits your actual package files and project structure to ensure instructions are grounded in your real-world scripts and file patterns, preventing the LLM from hallucinating incorrect commands.

    The skill is specifically optimized for Cursor (supporting both legacy .cursorrules and modern .mdc scoped rules) and Claude Code (supporting CLAUDE.md and .claude/rules/ configurations).

    Yes, the skill includes a dedicated migration feature that parses your existing root .cursorrules file and splits it into organized, scoped .mdc files with the correct YAML frontmatter and file globs.

    No, the generated rules are purely instructional context for the model; they do not require additional plugins or the installation of external security enforcement tools.

    The skill performs a 'Command Verification' step where it cross-references any suggested terminal commands against your project's actual build scripts and binaries to ensure they are executable.

    The refinement process is designed to merge new, evidence-based rules with your existing documentation, preserving your manual edits while highlighting and resolving any logic contradictions.

    New in Prompt & Skill Engineering