---

name: "do-test-recommendations" description: "Generate strategic test recommendations based on test-coverage-audit output. Produces prioritized test plan with specific recommendations for each gap. Use after running test-coverage-audit."

Test Recommendations

Transform test coverage audit findings into actionable, prioritized recommendations.

Input: Output from test-coverage-audit skill Output: Strategic test plan with specific recommendations

Prerequisites

This skill requires a completed test-coverage-audit. If no audit exists:

1. Run test-coverage-audit first
2. Or provide audit data in the prompt

Process

Step 1: Parse Audit Findings

Extract from audit report:

• Architecture type (determines testing strategy)
• Complexity sources (what needs testing)
• Gap analysis (what's missing)
• Quality issues (what's wrong with existing tests)

Step 2: Apply Strategic Framework

Testing Strategy by Architecture

Architecture	Primary Strategy	Key Considerations
Monolith	Test pyramid	Unit > Integration > E2E
Microservices	Contract-first	Consumer contracts, service mocks
Serverless	Integration-heavy	Function isolation, cold start
Real-time	Event-driven	Message ordering, replay
Data Pipeline	Data validation	Schema, lineage, idempotency

Read appropriate strategy reference:

• strategies/pyramid.md - Classic test pyramid
• strategies/contract-testing.md - Service contracts
• strategies/event-driven.md - Async/message systems
• strategies/data-validation.md - Data pipelines
• strategies/property-based.md - Property/fuzz testing

Gap Prioritization Matrix

For each gap identified in audit, score:

Factor	P0	P1	P2
Impact	Data loss, security, money	UX degradation, partial failures	Minor annoyance
Likelihood	Common path, no fallback	Edge case, has fallback	Rare, recoverable
Complexity	Simple to test	Moderate effort	Complex setup

Read: prioritization/risk-matrix.md

Step 3: Generate Recommendations

For each gap, produce:

### Recommendation: [Gap Name]

**Priority**: P0/P1/P2
**Gap**: [What's missing]
**Risk**: [What could go wrong]
**Recommendation**: [What to test and how]

#### Test Type
[Unit | Integration | E2E | Contract | Property]

#### Test Level
[Where in pyramid this belongs and why]

#### Approach
[Specific testing approach]

#### Framework/Tools
[Specific tools for the language/scenario]

#### Example Test Structure
```[language]
// Pseudocode or actual test structure

Effort Estimate

[Low | Medium | High] - [Why]

Dependencies

[What must exist before this test can be written]

### Step 4: Sequence Recommendations

Order recommendations by:

1. **Foundation first**: Tests that enable other tests
2. **P0 gaps**: Critical business/security risks
3. **Quick wins**: High value, low effort
4. **P1 gaps**: Important but not critical
5. **P2 gaps**: Nice to have

### Step 5: Identify Testability Blockers

Flag code that cannot be tested as-is:

```markdown
### Testability Blockers

| Area | Blocker | Refactoring Needed |
|------|---------|-------------------|
| PaymentService | Hard-coded Stripe client | Extract interface |
| UserAuth | Global state | Dependency injection |
| EmailSender | No interface | Create sendable interface |

**Note**: These blockers should be addressed before writing tests.
For refactoring recommendations, see `test-implementation-plan` skill.

---

Output Format

# Test Recommendations Report
**Project**: [name]
**Date**: [date]
**Based on**: Test Coverage Audit from [date]

## Executive Summary

**Architecture**: [type]
**Strategy**: [recommended approach]
**Total Gaps**: [n]
**Critical (P0)**: [n]
**Testability Blockers**: [n]

---

## Recommended Testing Strategy

[1-2 paragraph summary of the overall approach]

---

## P0 Recommendations (Critical)

### 1. [Gap Name]
[Full recommendation template]

### 2. [Gap Name]
[Full recommendation template]

---

## P1 Recommendations (Important)

### 3. [Gap Name]
[Full recommendation template]

---

## P2 Recommendations (Nice to Have)

### 4. [Gap Name]
[Full recommendation template]

---

## Testability Blockers

| Priority | Area | Blocker | Effort |
|----------|------|---------|--------|
| High | PaymentService | Hard-coded client | Medium |
| Medium | UserAuth | Global state | High |

---

## Implementation Sequence

Phase 1: Foundation (Week 1) ├── Fix testability blocker: PaymentService ├── P0-1: Payment error handling tests └── P0-2: Auth flow E2E

Phase 2: Core Coverage (Week 2) ├── P1-1: Webhook tests ├── P1-2: Cache invalidation └── P1-3: Database constraint tests

Phase 3: Hardening (Week 3+) ├── P2-1: Config validation ├── P2-2: Log output tests └── Flaky test cleanup

---

## Quick Wins

Tests that provide high value with minimal effort:

| Test | Value | Effort | ROI |
|------|-------|--------|-----|
| Auth E2E | High | Low | Excellent |
| Input validation | High | Low | Excellent |
| Error response format | Medium | Low | Good |

---

## Next Steps

1. **Immediate**: Address P0 gaps
2. **This sprint**: Fix testability blockers + P1 gaps
3. **Ongoing**: P2 gaps + test quality improvements
4. **For detailed implementation plan**: Run `test-implementation-plan` skill

---

Reference Documents

Strategies

Strategy	Reference
Classic test pyramid	strategies/pyramid.md
Contract testing	strategies/contract-testing.md
Event-driven testing	strategies/event-driven.md
Data validation testing	strategies/data-validation.md
Property-based testing	strategies/property-based.md

Prioritization

Topic	Reference
Risk matrix	prioritization/risk-matrix.md
Effort estimation	prioritization/effort-estimation.md
Quick wins	prioritization/quick-wins.md

---

Integration

This skill consumes output from test-coverage-audit and produces input for test-implementation-plan.

test-coverage-audit → test-recommendations → test-implementation-plan
      (What's wrong)      (What to do)           (How to do it)

Related Skills

Skill	Purpose
test-coverage-audit	Forensic analysis (run first)
test-implementation-plan	Execution plan with refactoring
do:add-tests	Write specific tests
do:tdd-workflow	Test-first development