Project Detail
GuardRail / VibeAudit
AI-native production-readiness platform for vibe-coded apps. Seven parallel security scanners (secrets, vulnerabilities, AST analysis, OSINT, compliance) with AI remediation and deployment intelligence.
Technical Summary
StackTypeScript + Python
Platformcli, web
RoleBackend Developer
TechnologiesTypeScript, Express, Python, MongoDB, Semgrep, Gitleaks, Trivy, OWASP Dependency-Check, OpenAI
ContextPersonal Project
Team Size1
Key Responsibilities
- Architected multi-scanner concurrent execution pipeline
- Built result normalization and scoring algorithms
- Integrated AI remediation with LLM APIs
- Created CLI tool with pre-push hook system
Project Insights
Built a multi-scanner security platform with concurrent execution, result normalization, and AI-powered remediation
Technical Concepts
- Seven parallel scanner modules for comprehensive security analysis
- Repository extraction, hashing, and caching for efficiency
- Vibe Score calculation based on production-readiness criteria
- AI-powered remediation with contextual fix suggestions
Architecture & Features
- Concurrent scanning with normalization and deduplication
- Severity correlation across different scanner outputs
- Git pre-push hooks for early security feedback
- Scan history tracking and trend analysis
Key Learnings
- Security tooling must be fast to fit developer workflows
- Normalization is critical when aggregating multiple scanner outputs
- AI remediation must provide actionable, context-aware suggestions