X.com Steve Reiner @stevereiner LinkedIn Steve Reiner LinkedIn Posts
Repost of my LinkedIn post from July 6, 2026 (original LinkedIn post)
The Flexible GraphRAG app now supports optional visual Langflow flows to customize the ingest pipeline and search / ai-query flows used “under the covers” of the UI (OpenRAG like). Also have configurable “Flexible” Langflow components to use in agentic apps or for RAG / GraphRAG. These include data source (13 sources ), document processor (Docling / LlamaParse), text splitter/chunking, KG extraction supporting ontologies for both pg and rdf dbs, vector (10 dbs), search (3 dbs), property graph (15 dbs). rdf graph (4 dbs), hybrid search, ai query / chat
Github: https://github.com/stevereiner/flexible-graphrag
Update: works with the incremental auto update sync
From my follow-up LinkedIn post from July 8, 2026 (original LinkedIn post)
I forgot to mention that the Flexible GraphRAG incremental auto update sync system works with the customizable visual ingestion pipeline flow (if ENABLE_LANGFLOW_FLOWS is configured). The flow is used for initial ingests, adds, and updates. The Langflow components are Python based and use the Flexible GraphRAG Python "framework“. Auto update sync with the Hyland Alfresco data source is tested and works with the ingest flow. These should work too: S3, Google Drive, OneDrive, SharePoint, Box, and Local Filesystem. For GCS and Azure Blob Storage, I have a fix coming soon. Postgres is used to manage the update system. Github: https://github.com/stevereiner/flexible-graphrag
