V

vector-search

A persistent-memory AI agent with cryptographically signed, hash-chained, tamper-evident memory. Pre-commit quality scoring quarantines prompt-injection; protected zones guard core records. Scans its own history for recurring gaps and proposes improvements for human review. Provider-agnostic (Claude, GPT, Gemini, DeepSeek, OpenRouter, Ollama).

PaveDB — A lightweight, pluggable vector search microservice with built-in document ingestion and deep observability.

Upload → chunk → index (with metadata) → search via REST and CLI, scoped by multi-tenant collections.

Multi-source RAG pipeline with hybrid vector + keyword retrieval, LLM-powered concept knowledge graph, adaptive search weighting, and evaluation framework.