A collection of Agent Skills for Postgres, curated from practitioners with years of real-world experience building and scaling applications on Postgres.
npx skills add neondatabase/postgres-skillsSkills are folders of instructions, scripts, and resources that agents can discover and use to perform tasks more accurately. Once installed, skills are automatically invoked when the agent detects relevant tasks.
Each skill starts with a SKILL.md file that serves as the entry point, allowing agents to progressively discover more detail as needed.
skills/postgres-best-practices
Best practices and guidelines for working with Postgres. Covers schema design, indexing, query optimization, and common pitfalls.
Give every AI engineer instant access to the collected experience of a staff engineer who has seen what works and what fails in production Postgres. Once installed, these skills guide your agent toward proven patterns and away from common mistakes.
- Vendor-agnostic. The skills are for Postgres users, regardless of which provider or deployment they use.
- Curated contributors. Content comes from well-known, credible Postgres practitioners with demonstrated real-world expertise.
- Paid commissions. Contributors are compensated for their time to keep the quality bar high.
- Stas Kelvich — [REVIEWER] Cofounder of Neon and a long-time Postgres contributor. Deep experience with scaling and tuning Postgres for large production workloads.
- Jonathan Katz — [REVIEWER] PostgreSQL Core Team member and Principal Product Manager at AWS (Amazon RDS). Active in the PostgreSQL community for over 15 years: co-chair of PGConf US, Secretary of the US PostgreSQL Association, and a contributor to pgvector.
- [You] - If you have experience scaling Postgres and would like to contribute, please reach out.
We are looking for skill contributors with:
- Deep, proven experience helping a diverse set of applications build and scale on Postgres
- A track record of writing clearly about Postgres (blog posts, documentation, conference talks)
Skill contributions are paid. Open an issue or reach out to andy.hattemer@databricks.com if you're interested in contributing.
Skills must be written by humans. AI-generated content defeats the purpose: if the information can be generated by an AI on demand, there is no value in encoding it into a skill ahead of time.
The value of a skill comes from human judgment, accumulated experience, and insight that AI does not already have. That is what makes it worth installing.
AI tools may be used for reviewing and testing skill content, but not for authoring it.