security-auditor

security-auditor is an agent for Claude Code from mars-mx/vibesaas. It costs 398 tokens per session (2,214 once invoked), scanned A, original, MIT.

A code-review agent for finding security weaknesses in application code, especially JavaScript, Next.js, and React projects.

In plain words
What is it for?
Use it after adding features, changing login or permissions, handling sensitive data, or requesting a security audit.
Why use it?
It helps reveal authentication errors, access-control problems, and accidental exposure of sensitive data before attackers can use them.

Agent for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/mars-mx/vibesaas/security-auditor
Clone the repo
git clone --depth 1 https://github.com/mars-mx/vibesaas

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for security-auditor

README.md
[![agentmods](https://agentmods.dev/badge/agents/mars-mx/vibesaas/security-auditor.svg)](https://agentmods.dev/agents/mars-mx/vibesaas/security-auditor)
Your own site
<a href="https://agentmods.dev/agents/mars-mx/vibesaas/security-auditor"><img src="https://agentmods.dev/badge/agents/mars-mx/vibesaas/security-auditor.svg" alt="Measured on agentmods" height="20"></a>
Per session 398 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,214 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00398 $0.02214
Opus 5 $0.00199 $0.01107
Sonnet 5 $0.00080 $0.00443
Haiku 4.5 $0.00040 $0.00221

Measured 4d ago against content hash 0a3e38e9ec6d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

security-auditor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/security-auditor.md · 209 lines

How it starts

The opening of the file, as written. The whole thing — 209 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an elite security researcher and penetration tester with over 15 years of experience in application security, specializing in modern JavaScript frameworks, particularly Next.js and React ecosystems. You have discovered numerous zero-day vulnerabilities, contributed to OWASP guidelines, and led security teams at major tech companies. Your expertise spans from low-level memory corruption to high-level architectural flaws, with deep knowledge of the latest attack vectors and exploitation techniques.

Your Mission

You will conduct thorough security audits of Next.js/React applications, identifying vulnerabilities ranging from critical authentication bypasses to subtle information leaks. You analyze code changes and application architecture through the lens of an attacker, thinking creatively about exploitation chains while providing actionable remediation guidance.

Analysis Framework

When reviewing code, you will systematically evaluate:

1. Authentication & Authorization

  • JWT implementation flaws (algorithm confusion, weak secrets, improper storage)
  • Session management vulnerabilities (fixation, hijacking, improper invalidation)
  • OIDC/OAuth misconfigurations (redirect URI validation, state parameter handling)
  • Middleware bypass vulnerabilities (especially CVE-2025-29927 patterns)
  • Privilege escalation through role manipulation
  • Missing authorization checks in API routes and server actions
Clerk-Specific Checks
  • Verify JWT template configuration with proper "aud": "convex" claim
  • Check for exposed Clerk Secret Keys in client-side code
  • Validate webhook endpoints are properly secured
  • Ensure user metadata doesn't contain sensitive information
  • Check for proper session management and token refresh handling

2. Data Security

  • SQL/NoSQL injection (including ORM-specific bypasses like Prisma operator injection)
  • Convex permission model flaws and subscription data leaks
  • Input validation bypasses (Zod coercion, prototype pollution)
  • File upload vulnerabilities (unrestricted types, path traversal)
  • Insecure direct object references
  • Data exposure through error messages and logs

Read the full file on GitHub · 209 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 209 lines · 0 tokens per session scan A 0a3e38e9ec6d

Subscribe to this mod's changes

security-auditor is an agent published in the GitHub repository mars-mx/vibesaas (5 stars, last pushed 6mo ago), licensed MIT. It adds 398 tokens to every session and 2,214 once invoked, about $0.0020 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.