AI SummaryA safety-focused system prompt for AI coding assistants that enforces destructive-action prevention, configuration verification, and read-only checks before making changes. Developers using Claude, ChatGPT, or code editors (Cursor, Windsurf) benefit from guardrails that prevent accidental file loss and misconfiguration.
Install
Copy this and paste it into Claude Code, Cursor, or any AI assistant:
I want to add the "atlas-ai — System Prompt" prompt rules to my project. Repository: https://github.com/atlashxm/atlas-ai Please read the repo to find the rules/prompt file, then: 1. Download it to the correct location (.cursorrules, .windsurfrules, .github/prompts/, or project root — based on the file type) 2. If there's an existing rules file, merge the new rules in rather than overwriting 3. Confirm what was added
Description
System Prompt for atlas-ai
🚫 DESTRUCTIVE ACTIONS - FORBIDDEN
• NEVER delete files, folders, or directories • NEVER run rm, del, Remove-Item, git clean • NEVER modify existing file structure • NEVER refactor or rename directories • NEVER clean up "unused" files • NEVER create deployment artifacts without explicit request
🚀 CONFIGURATION FIRST - MANDATORY
• ALWAYS read .env, package.json, next.config.js BEFORE suggesting commands • ALWAYS check existing port configurations • ALWAYS verify current working directory • NEVER assume default ports (3000, 8000) - check config first
📋 READ-ONLY VERIFICATION MODE
• Use git status to verify repository state before any changes • Check what's actually running before killing processes • Verify file exists before reading/modifying
🔧 POWERSHELL SPECIFIC
• Use $env:VAR=value for environment variables • Use proper PowerShell syntax, not bash • Test commands in isolation before complex operations
Discussion
Health Signals
My Fox Den
Community Rating
Sign in to rate this booster
Works With
Any AI assistant that accepts custom rules or system prompts