Claude Skill

sql-analyst

Translate business questions into SQL, validate results, and explain assumptions. Use for analytics, reconciliations, and ad-hoc reporting.

LLM Mart · 0 points · 0 0 0

Virus-scanned Reviewed automatically before listing.

Full trust report

Download navinspire-ia-navin-navin_skills_sql-analyst-e9c73a3.zip · 0 KB

navinspire-ia/navin 31 4 forks AGPL-3.0 Updated 4d ago

Part of navinspire-ia/navin — 182 skills

Install

skills CLI npx skills add https://github.com/Navinspire-ia/navin/tree/main/navin/skills/sql-analyst

Claude Code claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install navinspire-ia-navin@llmmart

Git git clone https://github.com/Navinspire-ia/navin.git

The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole navinspire-ia/navin collection as a plugin from our marketplace. Git is the plain clone.

Skill manifest

SQL Analyst

Overview

Question → SQL → verified answer. Show the query and caveats.

Tools

Run SQL through the built-in db_query tool (SQLite files via sqlite_path, or named PostgreSQL/Supabase/MySQL/MariaDB connections from tools.database.connections). Results come back as a table with a row cap - refine queries instead of dumping tables.

Workflow

Clarify metrics definitions (what counts as “active”, timezone, etc.).
Inspect schema via database-explorer habits.
Write readable SQL (CTEs over nested mess when helpful).
Run with limits first; then aggregate.
Sanity-check totals against a second query or known benchmark.
Present: answer, SQL, assumptions, data freshness.

Rules

No UPDATE/DELETE unless explicitly requested and approved.
Watch for fan-out joins that inflate metrics.

Files (navin)

SKILL.md 1.1 KB

---
name: sql-analyst
description: Translate business questions into SQL, validate results, and explain assumptions. Use for analytics, reconciliations, and ad-hoc reporting.
metadata: {"navin":{"emoji":"📊","category":"data"}}
---

# SQL Analyst

## Overview

Question → SQL → verified answer. Show the query and caveats.

## Tools

Run SQL through the built-in `db_query` tool (SQLite files via `sqlite_path`, or named PostgreSQL/Supabase/MySQL/MariaDB connections from `tools.database.connections`). Results come back as a table with a row cap - refine queries instead of dumping tables.

## Workflow

1. Clarify metrics definitions (what counts as “active”, timezone, etc.).
2. Inspect schema via `database-explorer` habits.
3. Write readable SQL (CTEs over nested mess when helpful).
4. Run with limits first; then aggregate.
5. Sanity-check totals against a second query or known benchmark.
6. Present: answer, SQL, assumptions, data freshness.

## Rules

- No `UPDATE`/`DELETE` unless explicitly requested and approved.
- Watch for fan-out joins that inflate metrics.