mattgierhart/PRD-driven-context-engineering / temp/v0.5-requirements/prd-v05-technical-stack-selection
prd-v05-technical-stack-selection
VerifiedMake technology decisions for every product capability by discovering existing assets, evaluating vendor-aligned options, and categorizing as Reuse/Extend/Build/Buy/Integrate/Research during PRD v0.5 Red Team Review. Handles both greenfield and brownfield contexts. Triggers on "tech stack", "build or buy?", "what technologies?", "technical decisions", "what do we reuse?", "existing stack", "vendor constraint", "IBM-first", "what tools do we need?", "evaluate solutions", "select tech stack". Consumes FEA- (features), SCR- (screens), RISK- (constraints). Outputs TECH- entries with decisions, rationale, and cross-references. Feeds v0.6 Architecture Design.
| File | Kind | Size |
|---|---|---|
| LEARNINGS-LIGHTSTACK.md | markdown | 8.7 KB |
| SKILL.md | skill md | 14.0 KB |
| assets/evaluation-scorecard.md | markdown | 2.0 KB |
| assets/tech-reuse.md | markdown | 1.7 KB |
| assets/tech.md | markdown | 2.2 KB |
| references/brownfield.md | markdown | 3.6 KB |
| references/examples.md | markdown | 5.5 KB |
| references/ibm-products.md | markdown | 4.7 KB |
references/brownfield.md
View on GitHub# Brownfield Discovery Workflow
Load this reference when the user confirms that existing products share infrastructure with the new product being evaluated.
## Purpose
When a product family already exists, many technology decisions are pre-made. This workflow helps discover those existing assets before evaluating new options — preventing the anti-pattern of re-evaluating settled decisions.
## Discovery Interview
Ask these questions in order. Record answers in the discovery summary table.
### 1. Product Family Overview
- "What products currently exist in this family?"
- "Which products share infrastructure (clusters, databases, services)?"
- "Is there a shared component library or design system?"
- "Who operates the current infrastructure? Same team or different?"
### 2. Per-Area Asset Discovery
For each area, ask what exists and its current state:
| Area | Discovery Questions |
|------|-------------------|
| **Frontend** | What framework? What design system? Shared component library? Hosting? |
| **Backend** | What language/framework? Monolith or microservices? API patterns? |
| **Database** | What database(s)? Managed or self-hosted? Shared or per-product? |
| **Auth** | What auth system? SSO? Identity provider federation? |
| **Infrastructure** | What cloud? What orchestration (K8s, serverless)? What region? |
| **CI/CD** | What pipeline? Shared or per-repo? What's automated? |
| **Monitoring** | What observability? Error tracking? Alerting? |
| **Messaging** | Any queue or event system? What patterns? |
### 3. Constraint Discovery
- "Are there vendor constraints (e.g., must use IBM, AWS, Azure)?"
- "Are there compliance requirements (HIPAA, SOC2, FedRAMP)?"
- "Are there team skill constraints (only know Python, only know React)?"
- "Are there cost constraints (budget ceiling for new services)?"
## Categorization Framework
After discovery, categorize each capability area:
| Category | Criteria | Example |
|----------|----------|---------|
| **Reuse** | Existing asset covers the new product's need as-is | "Keycloak handles auth for enRoll. INTERPRET needs the same auth patterns." |
| **Extend** | Existing asset works but needs additions | "React exists but needs new domain-specific components for policy review." |
| **New** | No existing asset for this capability | "No document parsing service exists in the family." |
| **Replace** | Existing asset is wrong for the new product | "Current batch queue doesn't support the real-time pipeline INTERPRET needs." |
## Output Format
Create a summary table:
| Capability Area | Existing Asset | Source Product | Category | Notes |
|----------------|---------------|---------------|----------|-------|
| Frontend | React + Carbon | enRoll | Extend | Need new policy review components |
| Auth | Keycloak on IKS | enRoll | Reuse | Separate realm for INTERPRET |
| Database | PostgreSQL on IBM Cloud | enRoll | Reuse | May need additional schemas |
| Document parsing | None | — | New | Greenfield evaluation needed |
| AI/ML | None | — | New | watsonx.ai evaluation needed |
| Task queue | None | — | New | Pipeline orchestration needed |
## Key Principle
**Discover before evaluating.** The most expensive mistake in brownfield contexts is re-evaluating technology that's already deployed, operational, and working. If the existing asset covers 80% of the need, extend it — don't replace it with something marginally better.
The exception: if the existing asset is fundamentally wrong for the new product's requirements (different scale, different compliance, different performance profile), then Replace is the correct categorization. But justify the replacement explicitly.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation). | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK). | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |