OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Memory & RAG

Openkl

Openkl is a local-first, open-source Python knowledge and memory layer for AI agents, operated through a CLI. It stores user notes and ingested documents, organizes knowledge with an embedded Kùzu graph database, and supports vector and hybrid search with citation-oriented output.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Openkl — OpenViST project profile in Agent Memory & RAG.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Openkl when building or experimenting with an agent workflow that needs a local store for notes and documents, cross-source retrieval, or citation-oriented results. Its CLI is also suited to direct management of memory and graph data.

Who it is for

  • Developers building AI agent workflows that need to retrieve a user’s notes or reference documents.
  • People maintaining a local knowledge corpus for search across memories and ingested material.

Documented uses

  • Add, update, search and delete user-provided memory notes.
  • Ingest a local documents directory or an ArXiv paper into a grounding store.
  • Search across stored memory and grounding material.
  • Create, verify, open and list citations associated with search results.
  • Inspect the knowledge graph or run Cypher queries.
Before you choose

Requirements & limits

The repository is public and identifies the project as open source under the Apache License 2.0. The README documents installation using uv and describes the project as being at a very early development stage.

  • The documented setup uses uv and a Python project environment.
  • Canonical content is file-based, with derived structure in an embedded Kùzu graph database.
  • Search capabilities described include FastEmbed-backed vector search and hybrid search across memory and the grounding store.
  • The README documents citation creation and verification, as well as citation-ready JSON output for agent workflows.
  • The project is explicitly described as being at a very early development stage.

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 Openkl and initialize its database

Install the documented Python project dependencies and initialize the local database.

Before you start

  • Python project environment
  • uv package manager

Steps

  1. Install uv if it is not already available.

  2. From the Openkl repository, run the command below to install the project dependencies.

    Documented command
    uv sync

  3. Run ok doctor to initialize the database.

Expected outcome

The project dependencies are installed and the database initialization command has been run.

Guide limitations

  • The guide covers setup and initialization only, not an agent integration or search workflow.
  • The project is described as being at a very early development stage.
Documentation 1 ↗
Source-based practical guide

Add and search a memory note

Save a user-provided fact and retrieve it with a memory search.

Before you start

  • Openkl installed with its database initialized

Steps

  1. Add a note with ok mem add "API rate limit is 100 req/min".

  2. Search memory with ok mem search "rate limit".

Expected outcome

The note is stored as memory and the CLI is directed to search memory for the matching topic.

Guide limitations

  • The repository lists these commands as usage examples; it does not establish results for other queries or inputs.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
nowledge-co/OpenKL · Official homepage title · retained in its source language
How we review sourcesSuggest a correction