pipeshub-ai/pipeshub-ai

PipesHub is an open-source fully extensible AI context layer that unifies your business data for explainable enterprise search and agentic workflow automation.

What it solves

PipesHub provides a unified, governed context layer for enterprise AI. It solves the problem of connecting fragmented enterprise knowledge across an organization while maintaining strict access permissions and ensuring that AI-generated answers are trustworthy and explainable through precise citations.

How it works

It functions as an AI-native execution layer that integrates with over 30 enterprise connectors to index data in real-time or on a schedule. It combines vector similarity search (via Qdrant) with graph-backed retrieval (via Neo4j or ArangoDB) to capture complex relationships in data. The platform is "Bring Your Own Model," allowing users to self-host any LLM provider within their own VPC to ensure data privacy.

Who it’s for

It is designed for enterprises and developers who need to build RAG applications, AI agents, and enterprise search tools that are self-hostable, permission-aware, and grounded in their own internal data.

Highlights

  • Permission-Aware Search: Enforces source-level access controls so users only see authorized information.
  • Explainable AI: Delivers grounded answers with block-level citations to original documents.
  • No-Code Agent Builder: Allows the visual creation of AI agents and execution of actions across enterprise tools.
  • Multimodal Capabilities: Supports understanding of images, diagrams, scanned files, and voice interactions.
  • Extensible Architecture: Provides SDKs for Python, TypeScript, and Go, and supports the Model Context Protocol (MCP).
  • Broad Format Support: Handles PDF, Docx, XLSX, PPTX, CSV, Markdown, HTML, and Google Workspace formats.