Works with the AI Agents you already use
Frontend Security Auditor Pro - Find real vulnerabilities
Framework-aware frontend security audits with traceable evidence, fewer false positives, and actionable fixes.
$14.99
Frontend Security Auditor Pro - Find real vulnerabilities
Example session with this skill installed
Use Frontend Security Auditor Pro to perform a Standard security audit of this frontend repository.
Identify realistic vulnerabilities involving XSS, DOM manipulation, authentication, token handling, browser messaging, redirects, and security configuration.
Trace attacker-controlled inputs to sensitive operations, evaluate existing protections, and avoid reporting false positives.
For each finding, provide the affected file and line, severity, evidence status, potential impact, recommended fix, and regression-test suggestions.
Separate verified source-level weaknesses from issues requiring runtime or backend verification.
Do not modify files or perform active penetration testing.
- Read your context and instructions
- Compiled the frontend security auditor
- Generated the UI component
Frontend Security Audit Report
Mode: Standard
Scope: React/TypeScript frontend sample
Method: Static source review only
Runtime testing: Not performed
Executive Summary
| Category | Result |
|---|---|
| High severity | 1 |
| Medium severity | 0 |
| Confirmed source-level findings | 1 |
| Additional verification required | 2 |
The review identified a source-level HTML injection weakness involving attacker-controlled URL parameters. The finding includes the affected data flow, practical exploitation preconditions, and recommended remediation.
Finding FSA-001 — Unsafe HTML Rendering
Severity: High
Evidence status: Confirmed in source code; not runtime-tested
Location: src/components/SearchResults.tsx:4 (illustrative path)
Evidence:
const query = new URLSearchParams(location.search).get('q') ?? '';
return <div dangerouslySetInnerHTML={{ __html: query }} />;
Security analysis:
The URL query parameter is attacker-controlled and passed directly into an HTML-rendering sink without visible sanitization.
A crafted URL could cause the application to interpret untrusted content as HTML. Depending on browser policies and deployment configuration, this may enable DOM-based XSS.
Recommended remediation:
Use React's default escaped text rendering instead of dangerouslySetInnerHTML when rich HTML is unnecessary.
If HTML rendering is required, apply a vetted, context-appropriate sanitization policy.
Regression test:
Supply benign HTML-like input and verify that it is rendered as text rather than interpreted as markup.
Additional Verification Required
| Area | Reason |
|---|---|
| Backend authorization | Backend implementation not provided |
| Production CSP | Deployed HTTP response headers not inspected |
Recommended Next Steps
- Fix the confirmed source-level weakness.
- Add regression tests for untrusted input handling.
- Review the related authentication and browser trust boundaries.
- Verify production security headers with authorization.
Deliverables: Prioritized findings, source references, remediation guidance, verification gaps, and structured Markdown/JSON reporting.
The skill reports observable evidence and limitations rather than claiming that a static audit proves a deployment is secure.
SearchResults.tsx
TSX · React component
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Turn your AI coding agent into a systematic frontend security reviewer that finds meaningful vulnerabilities, not just suspicious code patterns.
Frontend Security Auditor Pro analyzes frontend source code to identify security weaknesses, trace potentially exploitable data flows, evaluate existing protections, and recommend actionable fixes.
Unlike simple pattern-matching scanners, it examines how untrusted inputs reach security-sensitive browser APIs, considers framework protections, and distinguishes evidence-backed findings from issues requiring further verification.
What it covers:
- Cross-site scripting (XSS), DOM injection, and unsafe HTML rendering
- Authentication flows, token storage, and session security
- Client-side authorization assumptions and sensitive data exposure
- CSRF, redirects, URL handling, and browser messaging
- Content Security Policy (CSP), browser security headers, and configuration
- Third-party dependencies, supply-chain risks, and build settings
- Framework-specific security patterns across React, Next.js, Vue, Nuxt, Angular, Svelte, and vanilla JavaScript/TypeScript
Choose your audit depth:
- Quick: Focused assessment of high-risk attack surfaces and common vulnerabilities.
- Standard: Comprehensive source-level review with evidence-based findings and prioritized remediation.
- Deep: Extended analysis of complex data flows, trust boundaries, and regression-test recommendations.
What you receive:
A structured security report with severity ratings, evidence classifications, affected files and lines, exploitation preconditions, recommended fixes, and regression-test suggestions. Markdown and machine-readable JSON reporting are supported.
The skill is read-only by default and designed for authorized source-code reviews. It does not require credentials, external services, or live penetration testing.
Ideal for developers, engineering teams, security reviewers, and anyone reviewing AI-generated frontend applications before deployment.
Find vulnerabilities worth fixing. Understand why they matter. Know how to prevent them from returning.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.
- Static analysis only by default. The skill reviews source code and available configuration. It does not automatically perform penetration testing or browser-based exploitation.
- No guarantee of complete vulnerability detection. Findings depend on available source code, agent capabilities, application complexity, and audit scope.
- Backend security cannot be fully verified from frontend code. Server-side authorization, authentication enforcement, and API security may require separate backend review.
- Production configuration may differ. Deployed security headers, CSP policies, cookie attributes, and infrastructure settings cannot be confirmed without appropriate runtime evidence.
- Dependency vulnerabilities are not automatically verified. The skill can review dependency configuration, but identifying currently disclosed vulnerabilities requires up-to-date advisory information or an authorized dependency audit.
- The bundled discovery helper is heuristic. It identifies suspicious code patterns for further review, not confirmed vulnerabilities. Complex data flows require agent analysis.
- No automatic fixes. The default behaviour is read-only. Code modifications require explicit authorization.
- No automatic external data collection. The packaged helper operates locally, but data handling by the AI agent itself depends on the agent provider and environment.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 27 days ago
Permissions
File Scopes
This skill requires read access to authorized frontend source code and relevant project configuration files. Terminal access is used only when running the optional, read-only Python discovery helper. The helper uses the Python standard library, performs no network requests, installs no dependencies, and does not modify project files. The skill does not require external accounts, API keys, environment variables, or network access. Security reports are delivered through the agent's normal response. Source modifications, saved report files, additional command execution, or active security testing require separate authorization. The bundled helper does not upload source code or expose secret values. However, the hosting AI agent may process project content according to its own privacy and data-handling policies. Only files within the authorized audit scope should be accessed.