# Testing Strategy

> Prioritizes tests based on ROI and critical user journeys. Use when: deciding test coverage, selecting test types, or budgeting effort.

- **Type:** Skill
- **Install:** `agentstack add skill-shinpr-agentic-code-testing-strategy`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [shinpr](https://agentstack.voostack.com/s/shinpr)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [shinpr](https://github.com/shinpr)
- **Source:** https://github.com/shinpr/agentic-code/tree/main/.agents/skills/testing-strategy

## Install

```sh
agentstack add skill-shinpr-agentic-code-testing-strategy
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Test Strategy: ROI-Based Selection

## Core Principle: Maximum Coverage, Minimum Tests

**Philosophy**: 10 reliable tests > 100 unmaintained tests

Quality over quantity - focus resources on high-value tests that provide maximum coverage with minimum maintenance burden.

## ROI Calculation Framework

### ROI Formula

```
ROI Score = (Business Value × User Frequency + Legal Requirement × 10 + Defect Detection)
            / (Creation Cost + Execution Cost + Maintenance Cost)
```

### Value Components

**Business Value** (0-10 scale):
- 10: Revenue-critical (payment processing, checkout)
- 8-9: Core business features (user registration, data persistence)
- 5-7: Important secondary features (search, filtering)
- 2-4: Nice-to-have features (UI enhancements)
- 0-1: Cosmetic features

**User Frequency** (0-10 scale):
- 10: Every user, every session (authentication)
- 8-9: >80% of users regularly
- 5-7: 50-80% of users occasionally
- 2-4: 80% of users)
   - Core CRUD operations for primary entities
   - Critical business workflows
   - Data integrity for primary data models

**Budget Exception**: Critical User Journeys may exceed standard budget limits with explicit justification.

## Test Selection Guidelines

### Selection Thresholds

**Integration Tests**:
- ROI > 3.0: Strong candidate
- ROI 1.5-3.0: Consider based on available budget
- ROI  2.0: Strong candidate
- ROI 1.0-2.0: Consider if Critical User Journey
- ROI 3.0 for integration, >1.5 for E2E

3. **Budget Utilization**: Actual tests / Budget limit
   - Target: 80-100% (full utilization of valuable test slots)

4. **Defect Detection Rate**: Bugs caught / Total tests
   - Track over time to validate ROI predictions

### Calibration

Periodically review:
- Are high-ROI tests actually catching bugs?
- Are cost estimates accurate?
- Do business value ratings align with stakeholder priorities?

**Adjust formula weights based on empirical data**

## Anti-Patterns to Avoid

❌ **Gaming the System**:
- Inflating business value scores to justify favorite tests
- Ignoring ROI when it contradicts intuition
- Cherry-picking ROI calculation only for preferred tests

✅ **Proper Usage**:
- Apply ROI calculation consistently to all candidates
- Document justification when overriding ROI decisions
- Use empirical data to calibrate scores over time

❌ **Analysis Paralysis**:
- Spending excessive time on precise ROI calculations
- Debating single-point differences in scores
- Treating ROI as exact science rather than decision aid

✅ **Practical Application**:
- Use ROI for relative prioritization, not absolute precision
- Focus on order-of-magnitude differences (8.9 vs 0.34)
- Make quick decisions for obvious high/low ROI cases

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [shinpr](https://github.com/shinpr)
- **Source:** [shinpr/agentic-code](https://github.com/shinpr/agentic-code)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-shinpr-agentic-code-testing-strategy
- Seller: https://agentstack.voostack.com/s/shinpr
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.