Overview

Five practice areas to help you use AI agents safely on Veracity: securing credentials, regulated data, tenancy boundaries, code verification, and testing.

These five practices help you use AI agents safely when building on Veracity.An AI agent is a program (Cursor, GitHub Copilot, Claude Code, Codex, and similar tools) that uses an LLM to read your code, run commands, and call APIs on your behalf. When working with Veracity, where multi-tenancy, regulated maritime data, and OAuth credentials are all in scope, that broad access requires deliberate safeguards.

Practice areas

Before starting, configure the Veracity MCP server and follow your agent's setup guide to connect your agent to live Veracity documentation.