AI SummaryThis project includes deepsafe-scan, a preflight security scanner for AI agent environments. When the user asks to audit, scan, or check security of their AI agent setup, skills, or MCP servers, run the scanner: For a full scan with LLM analysis (if OPENAIAPIKEY is set):
Install
Copy this and paste it into Claude Code, Cursor, or any AI assistant:
I want to add the "deepsafe-scan — Cursor Rules" prompt rules to my project. Repository: https://github.com/XiaoYiWeio/deepsafe-scan 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
Universal preflight security scanner for AI coding agents — Detects hooks injection, credential exfiltration & backdoors in .cursorrules, CLAUDE.md, AGENTS.md and more.
DeepSafe Scan — Cursor Integration
This project includes deepsafe-scan, a preflight security scanner for AI agent environments.
Security scanning
When the user asks to audit, scan, or check security of their AI agent setup, skills, or MCP servers, run the scanner: `bash python3 scripts/scan.py --modules posture,skill,memory,hooks --scan-dir . --no-llm --format markdown ` For a full scan with LLM analysis (if OPENAI_API_KEY is set): `bash python3 scripts/scan.py --modules posture,skill,memory,hooks,model --scan-dir . --format markdown `
When to suggest scanning
• User installs a new MCP server or Cursor extension • User clones a project with .cursorrules or agent config files • User asks about security of their AI setup • User reports unexpected AI behavior that might indicate injection
Findings
Present findings by severity (CRITICAL → HIGH → MEDIUM → LOW). For each finding, show: what the risk is, the evidence, and how to fix it. Offer to fix CRITICAL and HIGH issues immediately.
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