/should - Best Practice Guidance

Get best practice recommendations, decision guidance, and tradeoff analysis.

System Prompt

⚠️ EXECUTION DIRECTIVE: When the user invokes this command, you MUST:

1. IMMEDIATELY execute - no questions, no explanations first
2. Parse the query to identify the decision point
3. Analyze tradeoffs using ADRs, standards, and best practices
4. Provide clear recommendation with reasoning

DO NOT:

• Give wishy-washy "it depends" without concrete guidance
• Skip the WHY behind recommendations

The user invoking the command IS the confirmation.

---

Usage

/should <decision-query>

Query Patterns

Query Pattern	Purpose	Sources
/should I use X or Y	Compare options	ADRs, standards, best practices
/should we refactor X	Refactoring decision	Code metrics, debt analysis
/should this be X	Design decision	Architecture patterns
/should I add X	Feature decision	Scope, complexity analysis
/should we test X	Testing decision	Risk analysis, coverage

Execution Steps

Step 1: Parse Decision Point

# Identify the decision being asked
# Common patterns: "X or Y", "refactor", "add", "change", "test"

Step 2: Gather Context

# Check relevant ADRs
grep -r "$TOPIC" internal/architecture/adrs/*.md

# Check standards
grep -r "$TOPIC" coditect-core-standards/*.md

# Check existing patterns
python3 scripts/component-indexer.py --search "$TOPIC"

Step 3: Analyze Tradeoffs

# For each option, identify:
# - Pros (benefits, alignment with standards)
# - Cons (risks, complexity, maintenance)
# - Precedent (how similar decisions were made)

Step 4: Provide Recommendation

# Clear YES/NO with reasoning
# Include: recommendation + why + alternatives + risks

Response Format

┌─────────────────────────────────────────────────────────────────┐
│  /should: <query>                                               │
├─────────────────────────────────────────────────────────────────┤
│  RECOMMENDATION                                                 │
│  ──────────────                                                 │
│                                                                 │
│  ✅ YES, you should... / ❌ NO, you should not...               │
│                                                                 │
│  REASONING                                                      │
│  ─────────                                                      │
│  <clear explanation of why>                                     │
│                                                                 │
│  TRADEOFFS                                                      │
│  ─────────                                                      │
│  Option A:                                                      │
│    ✓ Pro: <benefit>                                             │
│    ✗ Con: <drawback>                                            │
│                                                                 │
│  Option B:                                                      │
│    ✓ Pro: <benefit>                                             │
│    ✗ Con: <drawback>                                            │
│                                                                 │
│  PRECEDENT                                                      │
│  ─────────                                                      │
│  ADR-XXX: <relevant decision>                                   │
│  Standard: <relevant standard>                                  │
│                                                                 │
│  RISKS IF IGNORED                                               │
│  ───────────────                                                │
│  • <risk 1>                                                     │
│  • <risk 2>                                                     │
│                                                                 │
├─────────────────────────────────────────────────────────────────┤
│  Confidence: HIGH/MEDIUM/LOW                                    │
└─────────────────────────────────────────────────────────────────┘

Examples

Compare Options

/should I use SQLite or PostgreSQL for this feature

Output:

✅ YES, you should use PostgreSQL

REASONING:
The four-tier database architecture (ADR-118) establishes PostgreSQL
for production workloads. SQLite is only for local caching (Tier 3).

TRADEOFFS:
PostgreSQL:
  ✓ Pro: Production-ready, concurrent access, full SQL
  ✗ Con: Requires server setup, more complex
SQLite:
  ✓ Pro: Zero config, file-based, simple
  ✗ Con: Single-writer, limited concurrency

PRECEDENT:
ADR-118: Four-Tier Database Architecture

Refactoring Decision

/should we refactor the authentication module

Output:

✅ YES, you should refactor

REASONING:
Code metrics show:
- Cyclomatic complexity: 45 (threshold: 20)
- 3 duplicate implementations found
- No unit tests covering edge cases

RISKS IF IGNORED:
• Security vulnerabilities from unmaintained code
• Increased bug rate from complexity

Testing Decision

/should we add e2e tests for this PR

Output:

❌ NO, not for this PR

REASONING:
This is a small utility function with no UI changes. Unit tests
are sufficient. E2E tests would add CI time without proportional
coverage benefit.

ALTERNATIVE:
Add unit tests covering edge cases instead.

Related Commands

Command	Purpose
/can <action>	Check if something is possible
/how <task>	Get step-by-step guidance
/why <decision>	Understand existing decisions
/which <task>	Find the right tool/agent
/is <state>	Check current state

Success Output

✅ COMMAND COMPLETE: /should
Query: <query>
Recommendation: YES/NO
Confidence: HIGH/MEDIUM/LOW
Sources: <ADRs, standards, analysis>

Completion Checklist

Before marking complete:

• Decision point identified
• Tradeoffs analyzed
• Clear recommendation given
• Reasoning provided
• Precedent cited (if exists)
• Risks identified

Failure Indicators

This command has FAILED if:

• ❌ No query provided
• ❌ No clear recommendation given
• ❌ No reasoning provided
• ❌ "It depends" without specifics

When NOT to Use

Do NOT use when:

• Need capability check (use /can)
• Need state verification (use /is)
• Need step-by-step instructions (use /how)

Principles

This command embodies:

• #6 Clear, Understandable - Definitive recommendations
• #9 Based on Facts - Uses ADRs, standards, metrics
• #8 User Control Maintained - User makes final decision

Full Standard: CODITECT-STANDARD-AUTOMATION.md

---

Version: 1.0.0 Created: 2026-02-04 Author: CODITECT Team Framework: 5W+H Question Framework