neon

Query, branch, and inspect Neon PostgreSQL databases via psql and the Neon CLI.

Updated May 3, 2026
One-click install
npx skills add https://github.com/Claudfather/clauDNA --skill neon-claudfather
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: neon
Source: https://github.com/Claudfather/clauDNA/tree/main/skills/neon
Command: npx skills add https://github.com/Claudfather/clauDNA --skill neon-claudfather

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Working with Neon PostgreSQL requires juggling two CLIs (psql and neon), discovering connection strings from environment files, and avoiding accidental writes to production — this Skill consolidates all of that into one safe, verb-based workflow. ## Core Features & Use Cases - Query mode: Run ad-hoc SQL and schema exploration against Neon databases, with production queries automatically wrapped in read-only transactions and mutating SQL gated behind explicit confirmation. - Branch mode: Create, list, delete, or reset copy-on-write database branches for safe experimentation, pre-migration testing, and disposable dev environments. - Info mode: Generate a database dashboard showing connection status, database size, table inventory, and branch overview. - Use Case: Before running a risky migration, ask the agent to create a claude/experiment branch, test the migration against the branch connection string, then delete the branch — production is never touched. ## Quick Start Ask the agent to list all tables with row counts in your Neon database, or to create a disposable branch for testing a schema change.

Frequently Asked Questions about neon

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run SQL queries against a Neon PostgreSQL database?

Use psql with the connection string discovered from DATABASE_URL or a .env file, for example psql "<DB_URL>" -c "SELECT ...". Production queries are wrapped in BEGIN TRANSACTION READ ONLY to prevent accidental writes.

How do I create a Neon database branch for testing?

Run neon branches create with --project-id, --org-id, and a branch name such as claude/experiment-<date>. Then fetch its connection string with neon connection-string and run experimental queries against that URL.

Can I run INSERT or UPDATE statements on a production Neon database?

No. Production connections stay wrapped in read-only transactions with no exceptions. To mutate data, point at a development connection or create a disposable Neon branch and run the mutating SQL there after an explicit confirmation.

Why does neon me hang instead of returning in headless environments?

A bare neon me triggers the interactive OAuth device-code flow, which blocks forever in unattended or tmux contexts. Probe with timeout 10 neon me or pass --api-key with a non-empty NEON_API_KEY instead.

Why don't psql meta-commands like \dt work with a connection URL?

psql meta-commands such as \dt and \d+ do not work with the -c flag when using a connection URL. Use a heredoc instead, or query information_schema and pg_stat_user_tables with plain SQL equivalents.

What are the limits of Neon free tier branching?

The Neon free tier allows up to 10 branches, so check the current branch count before creating new ones. Point-in-time branching is also limited by the project's history retention window, which varies by plan.