Agent Skills: Fifteen-Factor App Methodology

Fifteen-Factor App Methodology

Overview

The Fifteen-Factor App methodology extends the original Twelve-Factor App principles (created by Heroku in 2012) with three additional factors essential for modern cloud-native applications: API First, Telemetry, and Security.

This methodology provides architectural principles and guidelines for building software-as-a-service applications that are:

• Performant - Optimised for speed and efficiency
• Scalable - Designed for horizontal scaling without significant changes
• Manageable - Easy to deploy, monitor, and maintain
• Resilient - Robust against failures with graceful degradation

When to Apply This Methodology

Apply the Fifteen-Factor principles during:

1. Architecture Planning - When designing new applications or microservices
2. PRP/PRD Creation - When documenting technical requirements and specifications
3. Code Reviews - When evaluating whether implementations follow best practices
4. Migration Planning - When modernising legacy applications for cloud deployment
5. Technical Debt Assessment - When identifying architectural improvements

The Fifteen Factors at a Glance

| Factor | Principle | Key Concept | |--------|-----------|-------------| | I. Codebase | One codebase, many deploys | Single repo per app, version controlled | | II. Dependencies | Explicitly declare and isolate | No implicit system-wide packages | | III. Config | Store in environment | Never hardcode configuration | | IV. Backing Services | Treat as attached resources | Databases, caches are swappable resources | | V. Build, Release, Run | Strict separation | Immutable releases, no runtime changes | | VI. Processes | Stateless and share-nothing | Horizontal scaling, no sticky sessions | | VII. Port Binding | Export via port | Self-contained, no runtime injection | | VIII. Concurrency | Scale out via process model | Horizontal over vertical scaling | | IX. Disposability | Fast startup, graceful shutdown | Maximise robustness | | X. Dev/Prod Parity | Keep environments similar | Continuous deployment | | XI. Logs | Treat as event streams | Separate generation from processing | | XII. Admin Processes | Run as one-off processes | Same environment as app | | XIII. API First | Design contracts first | Enable parallel development | | XIV. Telemetry | Monitor everything | APM, health checks, domain metrics | | XV. Security | Authentication & Authorisation | RBAC, identity per request |

Architecture Checklist

When creating a PRP, PRD, or architecture plan, evaluate the design against each factor. Use this checklist for quick validation:

□ Single codebase in version control
□ All dependencies explicitly declared
□ Configuration externalised to environment
□ Backing services abstracted and swappable
□ Build, release, run stages separated
□ Stateless processes (no sticky sessions)
□ Services self-contained with port binding
□ Designed for horizontal scaling
□ Fast startup and graceful shutdown
□ Dev/staging/prod environments aligned
□ Logs streamed to external aggregator
□ Admin tasks automated and reproducible
□ API contracts defined before implementation
□ Telemetry: APM, health checks, metrics
□ Security: Authentication and authorisation

Resources

Detailed documentation for each factor is available in the references directory:

• references/overview.md - Complete factor summary with diagrams
• references/original-factors.md - Factors I-XII with implementation examples
• references/modern-extensions.md - Factors XIII-XV (API First, Telemetry, Security)
• references/setup-and-tools.md - Tooling recommendations and quick start

Read the appropriate reference file for detailed guidance on specific factors.