ktx

ktx is an open-source context layer that turns warehouse metadata and analytics documentation into reviewable files for data agents.

ktx is an open-source context layer for data agents that turns warehouse metadata, BI tool definitions, query history, documentation, and approved metric definitions into reviewable files. It builds and maintains a context layer, then serves that context to agents at runtime. It is designed for data teams using AI agents to query analytical databases accurately. The system ingests source evidence from databases, BI tools, modeling code, and documentation, then reconciles it into a readable wiki and an executable semantic layer. Agents connect via MCP and CLI to search context, compile semantic queries, and run read-only SQL.

ktx at a glance

Best for: Software engineer, Finance
Runs on: CLI, Slack, open source, self-hosted
Pricing: Free
Released: May 2026
Category: AI research tool
Website: docs.kaelio.com

Key features

Write git-based wiki pages and semantic-layer definitions
Ingest data from warehouses, BI tools, and query history
Serve approved measures, joins, and filters to agents
Search context and compile semantic queries via CLI and MCP
Run read-only SQL on connected databases
Explain metric provenance with wiki content
Repair context through YAML and Markdown diffs

Use it for

Generate SQL from approved metrics, joins, filters, and dimensions
Explain metric provenance with wiki content and source evidence
Repair context through reviewable YAML and Markdown diffs
Work alongside dbt, MetricFlow, LookML, and other tools

Alternatives to ktx

Jonex — Provides traceable, agent-ready knowledge from diverse enterprise data sources.
OpenViking — Unifies agent memory, knowledge RAG, and skills into a single directory structure.
OpenMetadata — Unifies technical context and business semantics into one searchable graph.
Hex — It combines AI agents with governance, context management, and warehouse connections.
Cube — It keeps metric definitions consistent across AI agents and human users.
WeKnora — Combines RAG, multi-step agents, and automatic Wiki generation in one platform.
Yuxi — Combines RAG, knowledge graphs, and multi-agent orchestration in a self-hosted workspace.
Evidence — It defines business intelligence as code, enabling collaboration and automation.
DBX — It combines lightweight size, open-source access, and built-in AI for over one hundred databases.
DB-GPT — Builds autonomous agents for complex data tasks with private LLM technology.
RAGFlow — Provides an integrated platform combining data processing with agent workflows.
Metabase — Query and visualize data from any database using plain English, SQL, or clicks.

All ktx alternatives