Coach Development Resources

These resources document the tools and practices AIBMM coaches use to build software with AI assistants. Each page is structured to be read by both humans and AI — paste any page URL into your AI assistant to teach it our development workflow.

AI Quick Start

Point your AI at this page URL. It will read the resource index and understand what tools are available. Then navigate to any specific resource page for detailed instructions the AI can follow to build similar tooling in your project.

AI Context: Resource Index

This is the AIBMM Coach Resources home page. It contains links to eight development resources that document how AIBMM coaches build software with AI assistants. Each sub-page contains detailed, structured instructions that an AI can follow to replicate these tools in a new project. The resources are: (1) Getting Started - step-by-step environment setup with Ubuntu Server, Docker, and Cursor IDE via SSH, (2) Recursive AI Tool Loop - a production pattern for self-terminating AI analysis loops with tool orchestration, (3) Bob the Builder - deployment automation script with commit hash deploy verification, (4) Planning System - multi-agent architecture planning with Architect/Build/Test agent roles, (5) Requirements Traceability - user story templates with US-XXX/TC-XXX ID conventions and a traceability matrix connecting requirements to tests to code, (6) Testing - multi-layer testing with Playwright E2E and a TypeScript API test runner, (7) Security Review - full quality and security ESLint config plus code security audit process, (8) Storing Secrets in the Database - how to store API keys in PostgreSQL with AES-256-GCM encryption, a settings library with caching and env fallback, admin API, and security hardening. Navigate to each sub-page for the full guide.

Getting Started

Set up a professional AI-powered development environment from scratch: Ubuntu Server, Docker, Cursor IDE via SSH, and your first app running in under an hour.

AI use: Step-by-step guide to create a remote dev server, connect Cursor via SSH, run apps in Docker, and configure the AIBMM coach workflow with AI rules and planning templates.

DevOps & Infrastructure

The complete production stack: DigitalOcean App Platform, Managed PostgreSQL, GitHub Actions CI/CD, GHCR, Docker, and DNS — with step-by-step setup instructions.

AI use: Full infrastructure guide: DigitalOcean App Platform for hosting, Managed PostgreSQL for the database, GHCR for container images, GitHub Actions for CI/CD, Docker for containerization, and DNS configuration. Includes architecture overview and links to each service.

Recursive AI Tool Loop

A production-proven pattern for AI systems that analyze data, decide what they need, fetch it, and refine their own answer — without human intervention.

AI use: Full architecture guide with TypeScript templates for building a self-terminating recursive AI loop: orchestrator, tool system, pre-enrichment, living document, and safety rails.

Coulee Tech Deep Dive →

Bob the Builder

Our interactive deployment and quality-check script. Handles dev deployment, production releases, pre-flight checks, and testing — all from one menu.

AI use: Learn how BobTheBuilder.sh works and how to create similar deployment automation scripts for any project.

Deployment

DevOps

Bash

Planning System

A multi-agent planning framework using Architect, Build Agent, and Test Agent roles. Structure complex projects so AI can execute tasks independently.

AI use: Full markdown templates for the Architect → Build Agent → Test Agent workflow. Copy these into your project to organize AI-driven development.

Architecture

AI Agents

Project Planning

Requirements Traceability

The missing layer between planning and code. Structured user stories with ID conventions and a traceability matrix that connects every requirement to its tests and implementation.

AI use: Set up a docs/usecases/ pattern with user story templates (US-XXX), test case IDs (TC-XXX), and a TRACEABILITY.md matrix. Establishes user stories as the canonical source of truth.

Requirements

Documentation

Traceability

Testing

Our testing stack: Playwright for E2E browser tests, a TypeScript API test runner, and how quick-check catches issues before they hit the build.

AI use: Understand the testing layers (type-check, lint, E2E, TypeScript API runner with tsx) and how to set up similar testing automation in any Next.js project.

Security Review

How we review code for security: ESLint security plugin, dependency auditing, input validation with Zod, and infrastructure audit patterns.

AI use: Step-by-step guide for AI to perform a security review of a codebase — from dependency scanning to code-level vulnerability checks.

Security

Code Review

Best Practices

Storing Secrets in the Database

How to store API keys and service credentials in PostgreSQL when environment variables aren't an option — with AES-256-GCM encryption and an honest security review.

AI use: Full implementation guide: encrypted app_settings table, settings library with caching and env fallback, admin API with role checks, and security hardening checklist.

How Coaches Use These Resources

1. Start a New Project

Open your AI assistant (Cursor, Claude, ChatGPT, etc.) in a new project workspace. Paste the URL of this page or a specific resource page into the conversation.

2. AI Reads the Guide

The AI reads the structured content and understands the tool or workflow. Each page includes both the “what” and the “how to build it” instructions.

3. Build Consistently

The AI applies our patterns to your project. Every coach ends up with similar tooling — deployment scripts, planning systems, test suites, and security processes.