skills/neondatabase/agent-skills/neon/Gen Agent Trust Hub

neon

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install official Node.js packages (neon, @neon/config, @neon/env, @neon/sdk, @neon/functions) and download skill extensions from the vendor's GitHub repository. It also fetches live documentation and configuration protocols from official domains including neon.com and claimable.neon.tech.
  • [COMMAND_EXECUTION]: The skill heavily utilizes the Neon CLI (neon) to perform administrative tasks such as initializing projects, managing database branches, and deploying serverless functions. These operations involve file system modifications (e.g., updating .env files and .neon configuration) and network communication with the Neon control plane.
  • [DYNAMIC_EXECUTION]: The skill implements Infrastructure as Code (IaC) using a neon.ts TypeScript file. This file contains configuration logic and environment variable declarations that are executed by the Neon CLI during the deployment process to reconcile the state of cloud resources.
  • [INDIRECT_PROMPT_INJECTION]: The skill exposes an ingestion surface by reading external documentation and querying logs from the cloud platform. While these sources are official vendor endpoints, they represent a path for external data to enter the agent's context. Ingestion points include documentation pages from neon.com and log records retrieved through neon logs query (specified in SKILL.md and references/logs-loki.md). Capability inventory includes extensive CLI operations, network requests, and local file management across all skill files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 08:20 PM
Security Audit — agent-trust-hub — neon