Ground answers in your docs
RAG Compliance Chatbot
Knowledge base Q&A with hybrid vector + keyword search, full audit trail per query, and policy enforcement out of the box.
Key Capabilities
Hybrid Search
Combine semantic search with keyword matching using smart result fusion for higher recall than either method alone.
Document Parsing
PDF, DOCX, Markdown, and HTML documents are parsed, chunked (configurable overlap), and embedded automatically.
Audit Trail
Every query, retrieved chunk, and generated answer is immutably logged with full traceability. Exportable for compliance.
Policy Enforcement
Compliance policies (approval chains, data classification) are evaluated before returning answers. Blocked responses logged.
Per-Module Knowledge
Each module can reference different knowledge bases with custom retrieval strategy, topK, and score thresholds.
Guardrailed Responses
Built-in guardrails (PII, toxicity, schema, regex, moderation, prompt injection, hallucination) scan every response.
How It Works
Upload Documents
Drag-drop PDFs, DOCX, or Markdown files. Documents are parsed, chunked, and embedded automatically.
Configure Retrieval
Choose vector, keyword, or hybrid search. Set topK, score threshold, and chunk overlap per knowledge base.
Ask Questions
Natural language queries are embedded and matched against your knowledge base. Context injected into LLM prompt.
Audit & Comply
Every interaction is logged. Export audit trails for GDPR or internal compliance reviews.
Technical Stack
Ready to build production
Self-host in minutes with Docker, or use the cloud. Either way, you own your data and your models.