OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

Sag

Sag is a Python-based open-source retrieval architecture and knowledge-base application for turning documents and data into searchable, connected, traceable knowledge.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Sag — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Sag when retrieval needs to combine semantic search with relationships among events and entities, while preserving a path back to the original evidence.

Who it is for

  • Individuals building a local knowledge base
  • People working with document and structured-data retrieval
  • Developers supporting AI assistants or agents with traceable evidence
  • Users seeking a self-hosted RAG and knowledge-management application

Documented uses

  • Ingest file and web sources into a structured knowledge base
  • Search globally or within selected sources
  • Trace retrieval results and citations to original document chunks
  • Inspect event, entity, and relationship associations
  • Provide grounded agent answers through documented integration interfaces
Before you choose

Requirements & limits

The repository is public and identified as open source under the MIT license. The technical profile lists Linux, Windows, macOS, CLI, and self-hosted environments.

  • Its retrieval model creates query-time dynamic hyperedges instead of maintaining a globally pre-built graph.
  • Original evidence remains the output boundary, allowing selected events to map back to source chunks for generation and citation.
  • The documented starting stack uses SQLite and LanceDB and does not require an external database.
  • Integration options include REST/OpenAPI, OpenAI-compatible chat, MCP, and the zleap-sag Python package.
  • The project is described as local-first and single-user, which is relevant when evaluating deployment scope.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up Sag as a local knowledge-base application

Prepare the documented local-first application stack for indexing and retrieving knowledge.

Before you start

  • A Python environment; the repository README identifies Python 3.11+.
  • Node 20+ is identified in the repository README.
  • The SAG repository at https://github.com/Zleap-AI/SAG.

Steps

  1. Obtain the SAG project from the GitHub repository.

  2. Use the documented Python 3.11+ and Node 20+ environment.

  3. Use the local-first SQLite and LanceDB starting stack, which requires no external database.

  4. Proceed to the repository's User Guide or Developer Guide for the documented application workflow.

Expected outcome

A local-first SAG knowledge-base setup using SQLite and LanceDB as the documented starting storage components.

Guide limitations

  • These steps cover the documented setup direction, not a complete execution or deployment procedure.
  • The supplied documentation does not provide the exact installation commands or configuration steps.
  • The evidence does not establish hosted availability or account requirements.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
1 October 2026
Original website title
Zleap-AI/SAG · Official homepage title · retained in its source language
How we review sourcesSuggest a correction