Claude Code
AnthropicAnthropic's terminal agent: reads the repo, edits files, runs the tests it just broke.
Turns a business question into one reviewed query.
Recurring questions against a schema the team already understands
Writing to production tables — reads only, and a row limit is not optional
Given a schema and a question, it writes a single query with explicit column selection and a row limit, then explains the result. The rule that makes it safe is boring and effective: one query, no DDL, no writes, and the explanation must name the table and filter that produced the number.
Clone the repository, then copy the skill folder into your agent's skills directory — ~/.claude/skills/ for Claude Code. Skills are plain files: no build step, no daemon.
npx skills add sql-analystRead the SKILL.md before you install. A good skill states when it applies, what it must not do, and how to verify the result. A skill that only lists adjectives will change the agent's tone without improving its output — skip those.
Anthropic's terminal agent: reads the repo, edits files, runs the tests it just broke.
Copy generation scored against predicted performance.
Long-context writing and code work with a strong editing instinct.
Runs BigQuery queries and lists datasets through a scoped service account. The useful pattern is a question translated into one query with a row limit; the failure p…
Exposes Databricks catalog metadata and SQL execution for a workspace. It fits analytics teams that already live in notebooks; the risk is the same as any SQL tool —…
A community server that exposes Airtable bases, tables, and records as tools, including schema inspection and record creation. Useful when a non-engineering team kee…