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

Ktx

Ktx is an open-source TypeScript command-line context layer for data and analytics agents. It ingests warehouse metadata and business sources, maintains semantic-layer definitions and wiki context, and exposes search and MCP tools so agents can query analytical databases using approved metrics and relationships.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Ktx — 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 Ktx when agents need persistent, reusable business and warehouse context for recurring analytics queries; it sits on top of an analytical database and is most relevant when useful context sources are available to ingest.

Who it is for

  • Analytics and data teams using AI agents to query a SQL warehouse.
  • Teams that need agents to draw on approved metric definitions, joinable columns, and business knowledge spread across data models, BI tools, and documentation.

Documented uses

  • Provide agents with approved business metric definitions for warehouse queries.
  • Give agents table and column relationship context to help select joins.
  • Collect and maintain business knowledge used during analytics questions.
  • Expose semantic and wiki context to agents through CLI and MCP tools.
Before you choose

Requirements & limits

Open-source software under the Apache-2.0 license. The documented workflow uses the local CLI and requires the user's own LLM API keys or local agent sign-in; the README says Ktx adds no extra usage billing.

  • Ktx is a CLI context layer for an existing analytical database, rather than a warehouse or hosted analytics interface.
  • The README lists support for PostgreSQL, Snowflake, BigQuery, ClickHouse, MySQL, SQL Server, SQLite, DuckDB, Amazon Athena, and MongoDB.
  • Listed context sources include dbt, MetricFlow, LookML, Looker, Metabase, Sigma, Notion, and Google Drive.
  • Ktx serves agents through CLI and MCP and describes its SQL execution as read-only.
  • Setup configures providers, database connections, context sources, and agent integration in a local project.

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 a local Ktx project

Install Ktx and configure an initial project and agent integration.

Before you start

  • Node package manager (npm)
  • An LLM API key or local agent authentication
  • An analytical database and any context sources you intend to configure

Steps

  1. Install Ktx globally with the command below.

    Documented command
    npm install -g @kaelio/ktx

  2. Run ktx setup to create or resume a project and configure providers, connections, context building, and agent integration.

  3. Run ktx status to review project readiness.

  4. If status prints ktx mcp start --project-dir ..., run that command before opening the agent client.

Expected outcome

A configured local Ktx project with readiness information and agent integration; the MCP server can be started when indicated by status.

Guide limitations

  • This setup sequence does not document a specific agent query or validate a particular database integration.
  • Ktx requires the user's own LLM API keys or local agent sign-in.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
11 October 2026
Original website title
Kaelio/ktx · Official homepage title · retained in its source language
How we review sourcesSuggest a correction