AegisAI
VerifiedOpen-source AI Governance, Risk & Compliance (AI-GRC) platform. EU AI Act compliance engine, LLM Guard with prompt injection detection, and RAG-based regulatory intelligence.
Frameworks Covered
EU AI ActGDPRISO 42001
Services
compliance-managementllm-guardrag-intelligencerisk-classification
About AegisAI
AegisAI is a full-stack open-source AI-GRC platform that combines compliance management, LLM security, and regulatory intelligence into one platform. Designed for companies facing EU AI Act obligations.
Features
- EU AI Act risk classification (Minimal / Limited / High / Unacceptable)
- AI system registry with compliance dashboard
- Compliance document generation (Technical Docs, Risk Assessment, Conformity Declaration) with PDF export
- LLM Guard: real-time prompt injection detection using regex + DeBERTa-v3 ML classifier
- Per-user rate limiting and standalone Guard SDK
- RAG intelligence: natural language Q&A about EU AI Act, GDPR, ISO 42001
- SVG compliance badges for embedding in READMEs
- Bulk CSV import for AI system registration
- Guard scan GitHub Action for CI/CD integration
Tech Stack
- Frontend: React 18, TypeScript, Vite, Tailwind CSS, Zustand, TanStack Query
- Backend: Python 3.11, FastAPI, SQLAlchemy, PostgreSQL, Alembic
- ML: PyTorch, HuggingFace Transformers (DeBERTa-v3-small)
- RAG: LangChain, FAISS, OpenAI-compatible embeddings
- Infra: Docker Compose, Kubernetes, GitHub Actions CI
Loading reviews...