# Datus-ai/Datus-agent

The Future of Data Engineering — A CLI SQL client for the modern data stack, enabling AI-native context engineering for data.

Repository: https://github.com/Datus-ai/Datus-agent
Canonical: https://ross.abutalabs.com/products/datus-agent
Homepage: https://datus.ai/
Language: Python
License: NOASSERTION
License Family: other
Last push: 2026-08-26T13:46:40+00:00

## Health v2 (maintenance only)
Score: 84/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 96, longevity 30
- inputs: {"age_days": 425, "days_push": 7, "days_rel": 31, "gap_med": 12, "n_releases_24m": 16}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1613, forks 239 (observed 2026-08-28T04:05:11.044523+00:00)

## What it is
Datus is an open-source (Apache-2.0) AI data engineering agent delivered primarily as a Python CLI, with a VS Code/Cursor extension and optional cloud studio. It connects to data warehouses, catalogs, semantic layers, and BI tools, building an evolving context engine of schemas, metrics, and validated SQL to plan, generate, validate, and deploy data work.

## Use cases
- generate validated SQL from natural language against my warehouse
- build and maintain semantic models and metrics with AI
- connect Snowflake and query it with an AI agent from the terminal
- create an evolving context engine for my data stack
- generate data pipelines and ETL with an LLM agent
- self-host an AI SQL agent with my own model
- extract and reuse knowledge from SQL feedback and benchmarks

## When to choose
- you want an AI agent that plans, executes, and validates SQL end to end rather than a one-shot NL2SQL copilot
- you need a self-hosted, Apache-2.0 agent that works with your own warehouse and LLM
- you want persistent, evolvable context (schemas, metrics, validated SQL) shared across data tasks
- you prefer a CLI-first workflow with editor and MCP integrations

## When to avoid
- you need a simple GUI-only SQL client without AI features
- you require enterprise governance like SSO, RBAC, and audit logs on the free tier
- you want a fully mature, long-proven product — the project is young and fast-moving
- your team cannot cover its own LLM and warehouse compute costs

## Facets
- artifact type: cli-tool
- maturity: active
- function: agent-framework, llm-inference, rag, database, search-engine, cli, chatbot, mcp, etl
- domain: large-language-models, databases, analytics, developer-tools
- platform: python, cli, cross-platform, self-hosted
- tags: sql-agent, context-engine, nl2sql, semantic-layer, data-warehouse, metrics, vscode-extension, data-engineering, ai-agents

## Member repositories
- Datus-ai/Datus-agent (main) score 84

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:11.044523+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:51:04.578651+00:00, confidence not recorded.
  - readme: https://github.com/Datus-ai/Datus-agent (fetched 2026-08-28T04:05:11.044523+00:00, sha bf0b990acf9b)
  - homepage: https://datus.ai/ (fetched 2026-08-29T11:22:52.734455+00:00, sha 9884a40414f7)
  - site_page: https://docs.datus.ai (fetched 2026-08-29T11:22:52.748998+00:00, sha 36c6c3c2e4f9)
  - registry_pypi: https://pypi.org/pypi/datus-agent/json (fetched 2026-08-29T11:22:52.753399+00:00, sha dcf5c60a6b94)
  - site_page: https://datus.ai/integrations (fetched 2026-08-29T11:22:52.743737+00:00, sha 48d005558380)
  - site_page: https://datus.ai/pricing (fetched 2026-08-29T11:22:52.746505+00:00, sha e24922d4033a)
  - site_page: https://datus.ai/faq (fetched 2026-08-29T11:22:52.751040+00:00, sha 15d802f81182)
- Data as of 2026-08-30T08:39:29.467469+00:00.