nutim

Vega DB — Data skill for Claude Code

Data community

A distributed relational database built from scratch in Rust, designed for agentic workloads.

How to install Vega DB

This entry records only its repository, not the path inside it, so there is no exact command to give. Open nutim/vega-db and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Vega DB does

A distributed relational database built from scratch in Rust, designed for agentic workloads. It supports high-concurrency reads and writes, fast bulk ingestion of logs, memories, and datasets, easy schema introspection, and a built-in MCP server that lets Claude Code, Claude Desktop, and other MCP-compatible agents interact with the database direc

Alternatives in Data

HTML Anything — ✨ The agentic HTML editor — your local AI agent writes the HTML, you ship it 8.5k ★
Dummy Dataset — Generate realistic dummy datasets in CSV, JSON, or SQL 7.8k ★
Qsv — Blazing-fast Data-Wrangling toolkit 3554 17 1 3.6k ★

README

VegaDB

![VegaDB — Distributed SQL relational database in Rust](docs/assets/vegadb-banner.png)

*vega* (वेग) — Sanskrit for **speed, velocity**.

A **distributed relational database** written from scratch in Rust, designed for **agentic workloads**: many agents reading and writing concurrently, fast bulk ingest of logs/memories/datasets, easy schema introspection, and a **built-in MCP server** so LLM agents (Claude Code, Claude Desktop, or any MCP client) can use it as a tool.

flowchart TB
    subgraph clients[" "]
        shell["vegadb
(SQL shell)"]
        mcp["vegadb-mcp
(MCP tools)"]
        lib["your Rust
program"]
    end

    subgraph node["vegadb-server node"]
        engine["SQL engine
parse / plan"]
        raft["Raft
election · log replication · commit"]
        storage["MVCC storage
B+ trees · WAL · snapshots"]

        engine --> raft
        engine --> storage
        raft --> storage
    end

    shell -- "TCP / JSON" --> engine
    mcp -- "stdio / TCP" --> engine
    lib -- "embedded lib" --> engine
    raft <--> peers["other nodes"]

Features

Modern SQL engine
- Joins: INNER/LEFT/RIGHT/CROSS JOIN with ON/USING, table aliases, qualified columns (u.id), multi-way joins (hash join for equi-joins, nested loop otherwise)
- Aggregation: COUNT/SUM/AVG/MIN/MAX (incl. COUNT(DISTINCT x)), GROUP BY, HAVING
- Subqueries: scalar (SELECT ...), IN (SELECT ...), EXISTS (uncorrelated), derived tables FROM (SELECT ...) AS t, CTEs (WITH a AS (...), b AS (...))
- Set operations: UNION [ALL], INTERSECT, EXCEPT
- Expressions: CASE WHEN, CAST(x AS type), || concat, DISTINCT, ORDER BY on expressions/aliases/ordinals, scalar functions (COALESCE, NULLIF, LOWER, UPPER, LENGTH, SUBSTR, REPLACE, TRIM, CONCAT, ABS, ROUND, FLOOR, CEIL, RANDOM, NOW, CURRENT_TIMESTAMP, TYPEOF, ...)
- Modern DML: multi-row