A verifiable engineering knowledge engine that turns technical documents, CAD models, and equipment data into structured knowledge for humans and AI agents.
Verifiable engineering knowledge for humans and AI agents.
Turn engineering documents and CAD models into structured, searchable, and verifiable knowledge.
Technical documents · Engineering entities · Hybrid retrieval · Claim verification · MCP · Native STEP/STL
RagKernel is a verifiable engineering knowledge engine for building evidence-grounded systems over documents, CAD models, and equipment data. Additional engineering formats are planned behind the same ingestion contract.
While developing embedded systems and designing PCBs, I repeatedly searched through hundreds of pages of datasheets, reference manuals, and technical documents. Existing RAG systems could retrieve relevant text, but they often lost engineering context and the connection between an answer and its original evidence.
I also found that general-purpose AI assistants often struggled with engineering documents. They could produce plausible answers from datasheets, but those answers were not always grounded in the actual specification or supported by clear evidence.
Traditional RAG pipelines flatten engineering documents into text chunks. RagKernel preserves engineering structure, evidence, provenance, and geometry — enabling humans and AI agents to retrieve engineering knowledge that is traceable and verifiable.
Ingestion A unified ingestion interface routes each engineering source through the appropriate structured backend. Drop files in the web UI, a watched folder, or the CLI — idempotent, auto-indexed. - Documents — PDF (Docling + RapidOCR layout/table/OCR, real page numbers), DOCX / PPTX / HTML
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/v0id-byte--ragkernel.html)
Self-hosted documentation RAG pipeline that serves semantic search to LLM agents over the Model Context Protocol.
do-md/homekbLocal-first personal Markdown knowledge base with semantic search, AI Q&A, MCP, desktop/web clients, and a self-hostable zero-data-at-rest relay.
bybit-exchange/kaasTurn scattered notes, docs and transcripts into a queryable Markdown wiki — an LLM knowledge-base compiler with MCP access, no embeddings, self-hosted.
ozankasikci/global-agent-memoryLocal-first, project-aware memory MCP server for Claude Code, Codex, and other AI agents, with Obsidian and an owner dashboard.
fxd0h/Axelera-Voyager-Local-AssistantLocal AI assistant for the Axelera Voyager SDK and Metis AIPU — semantic RAG over the public Axelera sources + workflow skills, via MCP. No account, no cloud.
docouno/notariumFile-first knowledge base for humans and AI agents — self-hosted Markdown notes with a web editor and a built-in MCP server.
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.