Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add rules/neondatabase/ai-rules/neon-toolkitgit clone --depth 1 https://github.com/neondatabase/ai-rulesWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/rules/neondatabase/ai-rules/neon-toolkit)<a href="https://agentmods.dev/rules/neondatabase/ai-rules/neon-toolkit"><img src="https://agentmods.dev/badge/rules/neondatabase/ai-rules/neon-toolkit.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.02097 |
| Opus 5 | $0.00000 | $0.01048 |
| Sonnet 5 | $0.00000 | $0.00419 |
| Haiku 4.5 | $0.00000 | $0.00210 |
Grade A, and why
neon-toolkit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 213 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Neon Toolkit integration guidelines
This file provides comprehensive rules and best practices for interacting with Neon using the @neondatabase/toolkit. This toolkit is a terse client designed specifically for scenarios where a temporary, fully-functional Postgres database is needed quickly. As an AI agent, this is your primary tool for tasks requiring database interaction without manual setup.
The toolkit bundles the @neondatabase/api-client and the @neondatabase/serverless driver to streamline the entire lifecycle of a database: creation, querying, and deletion.
Neon Core Concepts
To use the toolkit effectively, you must understand Neon's resource hierarchy. While the toolkit simplifies this, knowing the concepts is crucial if you need to use the underlying apiClient.
| Concept | Description | Analogy/Purpose | Key Relationship |
|---|---|---|---|
| Organization | The highest-level container, managing billing, users, and multiple projects. | A GitHub Organization or a company's cloud account. | Contains one or more Projects. |
| Project | The primary container that contains all related database resources for a single application or service. | A Git repository or a top-level folder for an application. | Lives within an Organization (or a personal account). Contains Branches. |
| Branch | A lightweight, copy-on-write clone of a database's state at a specific point in time. | A git branch. Used for isolated development, testing, staging, or previews without duplicating storage costs. |
Belongs to a Project. Contains its own set of Databases and Roles, cloned from its parent. |
| Compute Endpoint | The actual running PostgreSQL instance that you connect to. It provides the CPU and RAM for processing queries. | The "server" or "engine" for your database. It can be started, suspended (scaled to zero), and resized. | Is attached to a single Branch. Your connection string points to a Compute Endpoint's hostname. |
| Database | A logical container for your data (tables, schemas, views) within a branch. It follows standard PostgreSQL conventions. | A single database within a PostgreSQL server instance. | Exists within a Branch. A branch can have multiple databases. |
| Role | A PostgreSQL role used for authentication (logging in) and authorization (permissions to access data). | A database user account with a username and password. | Belongs to a Branch. Roles from a parent branch are copied to child branches upon creation. |
| API Key | A secret token used to authenticate requests to the Neon API. Keys have different scopes (Personal, Organization, Project-scoped). | A password for programmatic access, allowing you to manage all other Neon resources. | Authenticates actions on Organizations, Projects, Branches, etc. |
| Operation | An asynchronous action performed by the Neon control plane, such as creating a branch or starting a compute. | A background job or task. Its status can be polled to know when an action is complete. | Associated with a Project and often a specific Branch or Endpoint. Essential for scripting API calls. |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 213 lines · 2,097 tokens per session scan A af4d759e2721
neon-toolkit is a cursor rule published in the GitHub repository neondatabase/ai-rules (86 stars, last pushed 3mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,097 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other cursor rules, from other repositories
adapter-features
Database-specific features must be implemented in the specialized adapter only. Base adapters (postgres, mysql, etc.) must remain database-agnostic.
unit-tests-tdd
TDD required for behavior changes; ≥80% package coverage on touched packages; unit-test conventions.
integration-tests
Human-readable integration test requests; helpers vs httptest; suites/ vs per-DB placement.
postgresql
PostgreSQL production rules. Safe migrations, parameterized queries, TIMESTAMPTZ, proper indexing strategy.
postgresql
This guide defines the definitive best practices for writing clean, performant, and maintainable PostgreSQL SQL, focusing on modern conventions and avoiding common pitfalls.
database-access
資料庫存取模式（Supabase client/server 分工）.