test-strategy-and-plan

name	Test Strategy and Plan
description	This skill should be used when the user asks to "create a test strategy", "define the testing approach", "plan our test levels", "set quality gates", "define test coverage targets", "create a test plan for this feature", or needs to define how an initiative will be tested end-to-end.
version	1.0.0

Test Strategy and Plan

Define a comprehensive testing approach that starts at requirements and covers every level of the test pyramid. Produce quality gates and tooling recommendations grounded in the system's risk profile.

Overview

A test strategy is not a list of tools — it is a risk-based decision about how much testing is enough and where to invest. This skill produces a strategy that the team can own and execute, not a document that collects dust.

Test Strategy Components

1. Risk-Based Test Scope

Identify the highest-risk areas requiring the heaviest testing:

Area	Risk Level	Reason	Test Investment
[Payment processing]	Critical	Financial data, compliance	High: unit + integration + E2E + security
[User authentication]	High	Security boundary	High: unit + integration + security
[Reporting export]	Medium	Data accuracy	Medium: unit + integration
[Static content pages]	Low	No logic	Low: smoke test only

2. Test Pyramid

Define the target test distribution:

        /\
       /E2E\         (10-20%): critical user journeys
      /------\
     / Integ. \      (30-40%): service integrations, API contracts
    /----------\
   / Unit Tests \    (50-60%): functions, business logic, edge cases
  /--------------\

Rationale for the chosen distribution (adjust based on system type):

• API-heavy systems: heavier integration layer
• UI-heavy systems: more E2E coverage
• Event-driven systems: add contract and consumer tests

3. Tooling Selection

Test Level	Tool	Reason
Unit	[Jest/pytest/JUnit]	[rationale]
Integration	[Supertest/pytest-httpx]	[rationale]
Contract	[Pact]	Consumer-driven contract tests
E2E	[Playwright/Cypress]	[rationale]
Performance	[k6/Locust]	[rationale]
Security	[OWASP ZAP/Burp]	[rationale]
Accessibility	[axe-core/Pa11y]	[rationale]

4. Quality Gates

Define pass/fail thresholds per pipeline stage:

Stage	Gate	Threshold	Action on Fail
Pre-merge	Unit tests	100% pass	Block PR
Pre-merge	Coverage	> 80% on new code	Block PR
Pre-merge	Linting	0 errors	Block PR
Pre-deploy (staging)	Integration tests	100% pass	Block deploy
Pre-deploy (staging)	E2E (smoke)	100% pass	Block deploy
Pre-deploy (production)	E2E (full suite)	100% pass	Block deploy
Pre-deploy (production)	Performance baseline	p95 < [target]ms	Block deploy

5. Shift-Left Test Activities

Embed testing earlier in the SDLC:

• At requirements: write acceptance criteria as Given/When/Then (BDD scenarios)
• At architecture: identify testability requirements (dependency injection, seams)
• At implementation: TDD for business logic, contract tests for integrations
• At CI: all tests run on every commit, not just pre-merge

6. Test Data Strategy

Define how test data is created and managed:

• Unit tests: inline test data and factories
• Integration tests: dedicated test database, seeded via fixtures
• E2E tests: synthetic test accounts, never production data
• Performance tests: anonymised production clone or generated realistic dataset

Test Plan Structure

For a specific feature or sprint, produce a test plan:

# Test Plan: [Feature Name]
Sprint: [N] | Author: VAMFI QA Lead

## Scope
In scope: [features covered]
Out of scope: [features not covered by this plan]

## Test Cases Summary
| Level | Count | Automated | Manual |
|---|---|---|---|
| Unit | [N] | [N] | 0 |
| Integration | [N] | [N] | 0 |
| E2E | [N] | [N] | [N] |

## Quality Gate Checklist
[ ] All unit tests passing
[ ] Coverage target met
[ ] Integration suite passing
[ ] E2E smoke suite passing
[ ] No P0 or P1 defects open

Additional Resources

• references/test-level-patterns.md — Detailed patterns per test level
• assets/test-strategy-template.md — Blank test strategy template