Skip to main content
skill CX Codex

security-visual-testing — Claude Code skill: what it does & install

security-visual-testing validates URLs, masks sensitive data on screenshots, and runs multi-viewport visual regression tests with WCAG accessibility audits.

MIT 115 installs ★ 533 updated 2026-10-25

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add proffesor-for-testing/agentic-qe --skill security-visual-testing -a claude-code
Open on GitHub

What security-visual-testing does

This skill combines visual layout regression with security inspection before writing screenshots to storage. It blocks dangerous protocols like javascript: or file: before navigation occurs. Browser tasks require the Vibium Go binary with WebDriver BiDi, installed through aqe init. Before saving image artifacts, the check-injection.js script scans DOM content for prompt-injection markers and aborts when issues appear.

The built-in detector spots credentials, payment details, and API keys inside forms. These elements receive intense blur or solid black masking prior to file export. Testing runs across four screen sizes at once: 320px mobile, 768px tablet, 1440px desktop, and 1920px wide layouts. Accessibility verification relies on axe-core rules targeting WCAG 2.1 AA compliance.

What it helps with

  • block risky URL protocols before browser navigation
  • mask credentials and API keys in test screenshots
  • run parallel visual diff tests across four screen sizes
  • audit web interfaces against WCAG 2.1 AA standards

Example prompts for the agent

  • Run a visual regression audit on https://example.com with security and WCAG checks.
  • Capture a page screenshot for https://example.com/profile with PII masking enabled.
  • Verify responsive layouts on https://example.com across mobile and desktop viewports.

How to install security-visual-testing

1Skills CLI (Vercel)

npx skills add proffesor-for-testing/agentic-qe --skill security-visual-testing -a claude-code

2Manually: put the files in place

~/.claude/skills/security-visual-testing/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/security-visual-testing/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about security-visual-testing

How do I install security-visual-testing in Claude Code?

Run npx skills add proffesor-for-testing/agentic-qe --skill security-visual-testing -a claude-code. Alternatively, copy the file to ~/.claude/skills/security-visual-testing/SKILL.md.

How does the skill prevent data leaks in test artifacts?

The scanner identifies credentials, credit card numbers, and API tokens before saving images. It obscures detected regions using blur or opaque boxes directly on the screenshot.

What browser setup is required before running tests?

The skill relies on the Vibium WebDriver BiDi engine installed via aqe init. It runs check-injection.js to block screenshots when prompt injection patterns are detected.

Source text

SKILL.md · MIT
---
name: security-visual-testing
description: "Security-first visual testing combining URL validation, PII detection, and visual regression with parallel viewport support. Use when testing web applications that handle sensitive data, need visual regression coverage, or require WCAG accessibility compliance."
category: specialized-testing
priority: high
tokenEstimate: 1400
agents: [qe-visual-tester, qe-security-scanner, qe-accessibility-auditor]
implementation_status: stable
optimization_version: 1.0
last_optimized: 2026-01-21
dependencies: [visual-testing-advanced, security-testing, accessibility-testing]
quick_reference_card: true
tags: [security, visual-testing, pii, accessibility, wcag, parallel-viewport, regression]
trust_tier: 3
validation:
  schema_path: schemas/output.json
  validator_path: scripts/validate-config.json
  eval_path: evals/security-visual-testing.yaml
---

# Security Visual Testing

## Browser engine

Uses the **qe-browser** fleet skill (`.claude/skills/qe-browser/`) for all browser automation. The Vibium engine (10MB Go binary, WebDriver BiDi) is installed automatically by `aqe init`. For security-visual workflows, qe-browser adds two things on top of stock visual testing: `check-injection.js` (scans page content for prompt-injection patterns before screenshots are stored) and `assert.js` (16 typed checks including `no_failed_requests` for detecting data-leak requests).

```bash
# Before storing any screenshot, scan the page
vibium go "$TARGET_URL"
vibium wait load
node .claude/skills/qe-browser/scripts/check-injection.js --include-hidden
INJ=$?
if [ $INJ -ne 0 ]; then
  echo "Prompt-injection findings — do NOT store screenshot"
  exit $INJ
fi
# Safe to proceed with visual-diff
node .claude/skills/qe-browser/scripts/visual-diff.js --name "${PAGE_NAME}"
```

<default_to_action>
When performing security-aware visual testing:
1. VALIDATE URLs before navigation (check for malicious patterns)
2. SCAN for PII before saving screenshots (mask sensitive data)
3. CAPTURE parallel viewports (mobile, tablet, desktop)
4. COMPARE against baselines (detect visual regressions)
5. AUDIT accessibility (WCAG 2.1 AA compliance)

**Quick Security-Visual Checklist:**
- URL validation (no javascript:, data:, file: schemes)
- PII detection (emails, phones, SSN, credit cards, API keys)
- Visual regression (diff threshold < 0.1%)
- Viewport coverage (320px, 768px, 1024px, 1440px)
- Accessibility score (> 90% axe-core pass rate)

**Critical Success Factors:**
- Always mask PII before storing screenshots
- Test across all target viewports in parallel
- Store baselines in version control
- Run accessibility audits on every visual change
</default_to_action>

## Quick Reference Card

### When to Use
| Scenario | Use This Skill? | Why |
|----------|-----------------|-----|
| Testing login pages | Yes | Contains PII (passwords, emails) |
| Visual regression suite | Yes | Parallel viewport + baseline comparison |
| Payment forms | Yes | Credit card data needs masking |
| Public marketing pages | Maybe | Only if sensitive data possible |
| API-only testing | No | Use security-testing skill instead |

### Key Capabilities

| Capability | Description | Performance |
|------------|-------------|-------------|
| URL Validation | Block malicious URLs before navigation | <5ms |
| PII Detection | Find 6+ types of sensitive data | <100ms |
| Parallel Viewports | Test 4 viewports simultaneously | 4x faster |
| Visual Regression | Pixel-diff with configurable threshold | <500ms |
| Accessibility Audit | WCAG 2.1 A/AA/AAA compliance | <2s |

---

## Workflows

### 1. Security Visual Audit (Full Pipeline)

```bash
# Run complete security + visual audit
aqe test visual-audit --url https://example.com --security --accessibility
```

**Steps:**
1. Validate URL security (block malicious schemes)
2. Scan page for security issues (XSS, injection patterns)
3. Capture screenshots across 4 viewports in parallel
4. Compare against stored baselines
5. Run accessibility audit (axe-core)
6. Generate consolidated report

### 2. PII-Safe Screenshot

```bash
# Capture screenshot with automatic PII masking
aqe screenshot --url https://example.com/profile --pii-safe
```

**PII Detection Patterns:**
- Email addresses: `[email protected]`
- Phone numbers: `+1-555-123-4567`
- Credit cards: `4111-1111-1111-1111`
- SSN: `123-45-6789`
- API keys: `sk_live_...`, `AKIA...`
- Passwords: Form fields with type="password"

**Masking Strategy:**
- Default: Blur with high intensity
- Options: `redact` (black box), `pixelate`, `blur`

### 3. Responsive Visual Audit

```bash
# Test visual consistency across viewports
aqe test responsive --url https://example.com --viewports mobile,tablet,desktop
```

**Default Viewports:**
| Name | Width | Height | Device |
|------|-------|--------|--------|
| mobile | 320px | 568px | iPhone SE |
| tablet | 768px | 1024px | iPad |
| desktop | 1440px | 900px | MacBook |
| wide | 1920px | 1080px | Full HD |

---

## Integration with AQE v3

### Using with BrowserSecurityScanner

```typescript
import { BrowserSecurityScanner } from '@agentic-qe/v3';

const scanner = new BrowserSecurityScanner(memory, {
  urlValidation: { enabled: true },
  piiDetection: { enabled: true, maskBeforeSave: true },
  parallelViewports: { maxConcurrent: 4 }
});

const result = await scanner.scanUrl('https://example.com', {
  viewports: ['mobile', 'tablet', 'desktop'],
  accessibility: true
});
```

### Using with TrajectoryAdapter

```typescript
import { TrajectoryAdapter } from '@agentic-qe/v3';

const adapter = new TrajectoryAdapter(memory);

// Record testing trajectory for learning
await adapter.startTrajectory('security-visual-test', {
  url: 'https://example.com',
  testType: 'security-visual'
});

// ... perform tests ...

await adapter.endTrajectory(trajectoryId, {
  success: true,
  piiFound: 3,
  visualRegressions: 0,
  accessibilityScore: 95
});
```

---

## Agent Coordination

### Memory Namespace
```
aqe/security-visual/
├── baselines/*          - Visual regression baselines
├── screenshots/*        - Captured screenshots (PII masked)
├── reports/*            - Audit reports
└── trajectories/*       - Learning trajectories
```

### Fleet Coordination
```typescript
const fleet = await FleetManager.coordinate({
  strategy: 'security-visual-audit',
  agents: [
    'qe-visual-tester',      // Visual regression
    'qe-security-scanner',   // URL/PII scanning
    'qe-accessibility-auditor' // WCAG compliance
  ],
  topology: 'parallel',
  maxConcurrent: 4
});
```

---

## Error Handling

| Error | Cause | Resolution |
|-------|-------|------------|
| `URL_BLOCKED` | Malicious URL pattern detected | Check URL, remove javascript:/data: |
| `PII_DETECTED` | Sensitive data found in screenshot | Enable masking or redact manually |
| `BASELINE_MISSING` | No baseline for comparison | Run with `--update-baseline` first |
| `VIEWPORT_TIMEOUT` | Browser didn't respond | Increase timeout or reduce parallel |
| `ACCESSIBILITY_FAILED` | WCAG violations found | Review violations, fix issues |

---

## Related Skills
- [visual-testing-advanced](../visual-testing-advanced/) - Pure visual testing without security
- [security-testing](../security-testing/) - Security testing without visual component
- [accessibility-testing](../accessibility-testing/) - Accessibility-only testing
- [qe-visual-accessibility](../qe-visual-accessibility/) - AQE v3 visual domain skill

---

## Performance Targets

| Metric | Target | Measured |
|--------|--------|----------|
| URL validation | <5ms | 2ms |
| PII detection | <100ms | 45ms |
| Single viewport capture | <2s | 1.2s |
| 4-viewport parallel | <3s | 2.1s |
| Visual diff | <500ms | 320ms |
| Accessibility audit | <2s | 1.5s |
| Full pipeline | <10s | 7.2s |

© proffesor-for-testing · MIT · proffesor-for-testing/agentic-qe

Similar skills

All skills for Testing and QA →

From the same set73

The whole set →
test-environment-management test-environment-management is a Claude Code skill for setting up test infrastructure and matching it to production. It covers environment setup, parity checks, and test infrastructure costs. CX Codex +1 117installs debug-loop debug-loop — a skill for step-by-step debugging of errors through real commands, not simulations. Forms hypotheses, validates them against live data, identifies root cause, and only then fixes the code. CX Codex +1 117installs shift-right-testing shift-right-testing guides agents through production testing using feature flags, canary rollouts, and chaos experiments. Useful when setting up progressive delivery or tracking SLOs. CX Codex +1 115installs qe-contract-testing qe-contract-testing — a skill for verifying REST, GraphQL, and event-driven API contracts with schema validation. Detects incompatible changes and ensures APIs won't break downstream clients. CX Codex +1 115installs enterprise-integration-testing Enterprise integration testing skill for SAP, middleware, and WMS setups. It coordinates cross-system checks across RFC, IDoc, OData, and ESB flows. CX Codex +1 113installs holistic-testing-pact holistic-testing-pact builds testing strategies around PACTS principles and holistic quadrants. It structures quality planning for classical, AI-assisted, and autonomous agent setups. CX Codex +1 111installs observability-testing-patterns observability-testing-patterns is a Claude Code skill with patterns for checking dashboards, alert thresholds, log aggregation, and APM tracing during monitoring tests. CX Codex +1 109installs sfdipot-product-factors The sfdipot-product-factors skill turns requirements into ranked test ideas using seven SFDIPOT dimensions. Use it to assess epics, user stories, or a test strategy. CX Codex +1 108installs

Project sponsors

Companies that keep this analytics open