secure-code-review

secure-code-review is a skill for Claude Code, Codex from zebbern/claude-code-guide. It costs 86 tokens per session (4,347 once invoked), scanned B, original, MIT.

A checklist for reviewing application code against common web security weaknesses from the OWASP Top 10, such as SQL injection, cross-site scripting, unsafe requests, and broken access controls. It includes examples of vulnerable code and suggested fixes.

In plain words
What is it for?
Use it to review files or code snippets, investigate specific vulnerabilities, or perform a broader security audit of a web application.
Why use it?
It gives a repeatable way to find security problems in source code before they reach production and explains how to correct them.

Skill for Claude CodeCodex

About the project

Claude Code Guide is a reference collection for configuring and using Claude Code, Anthropic’s command-line coding agent. Developers use it to learn commands, skills, agents, MCP, automation, security, integrations, and troubleshooting. Its catalogue add-ons provide many of the documented skills and agents.

zebbern/claude-code-guide · 4,600 stars · on GitHub

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 skills/zebbern/claude-code-guide/secure-code-review
Any agent
npx skills add zebbern/claude-code-guide --skill secure-code-review
Clone the repo
git clone --depth 1 https://github.com/zebbern/claude-code-guide

Made for: Claude Code, Codex.

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 secure-code-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/zebbern/claude-code-guide/secure-code-review.svg)](https://agentmods.dev/skills/zebbern/claude-code-guide/secure-code-review)
Your own site
<a href="https://agentmods.dev/skills/zebbern/claude-code-guide/secure-code-review"><img src="https://agentmods.dev/badge/skills/zebbern/claude-code-guide/secure-code-review.svg" alt="Measured on agentmods" height="20"></a>
Per session 86 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,347 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 3 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 $0.00086 $0.04347
Opus 5 $0.00043 $0.02174
Sonnet 5 $0.00017 $0.00869
Haiku 4.5 $0.00009 $0.00435

Measured yesterday against content hash d3c9fac00e49, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

secure-code-review scanned grade B with 3 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 yesterday.

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.

Cloud metadata endpointmediumServer-side request forgery

One request to 169.254.169.254 can return temporary IAM credentials.

- [ ] Is the cloud metadata address blocked (169.254.169.254)?

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

response = requests.get(url)

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

- [ ] Are there `os.system()` or `subprocess.call(shell=True)` calls that concatenate user input?
skills/secure-code-review/SKILL.md · 572 lines

How it starts

The opening of the file, as written. The whole thing — 572 lines — stays where its author put it; the contents beside it link to each section on GitHub.

OWASP Top 10 Code Security Review Checklist

A systematic security review based on the OWASP Top 10 (2021) standard. Each item includes: vulnerability description, typical vulnerable code, inspection checkpoints, and remediation examples. Designed for security-focused code review of web applications.

Usage

Provide the code files or code snippets to review, and specify which OWASP categories to check (or request a full review) to receive an item-by-item audit report.

Example prompts:

  • "Check this code for SQL injection risks"
  • "Run a full OWASP Top 10 security review on this project"
  • "Does this API endpoint have any SSRF vulnerabilities?"

Quick Reference

ID Category Key Check
A01 Broken Access Control Does every endpoint verify the current user's identity? Can users access others' data by changing IDs?
A02 Cryptographic Failures Are passwords hashed with bcrypt/argon2? Are secrets hardcoded?
A03 Injection String-concatenated SQL? shell=True? Unescaped template output?
A04 Insecure Design Is rate limiting in place? Can critical workflows be bypassed?
A05 Security Misconfiguration DEBUG enabled? Stack traces in error pages? Default credentials?
A06 Vulnerable Components Any CVEs from pip audit / npm audit?
A07 Authentication Failures Is JWT signature verified? Can tokens be revoked? Is MFA available?
A08 Integrity Failures Any pickle.loads deserializing untrusted data?
A09 Logging & Monitoring Failures Are plaintext passwords in logs? Are failed logins recorded?
A10 SSRF Are user-supplied URLs filtered against internal IPs?

Review Process SOP

Core principle: prefer false positives over missed true positives.

  1. Define scope — Identify the files, modules, or code snippets to review
  2. Full coverage check — Scan through A01-A10 sequentially. Every item must appear in the report (mark items with no findings as pass). The default behavior is to only report issues found — this process requires full coverage to ensure nothing is missed
  3. Risk classification — Label each finding:
    • RED High: Directly exploitable (RCE, SQL injection, SSRF reaching internal networks, plaintext password storage)
    • YELLOW Medium: Exploitable under specific conditions (missing rate limiting, weak password policy, static tokens)
    • GREEN Low: Defense-in-depth gap with no direct exploitation path (missing security headers, insufficient logging)
  4. Every finding must include ready-to-use fix code (actual code, not just a description). Reference specific file:line_number
  5. Output the review report — Use the template below, findings sorted by severity descending, with a prioritized remediation list at the end

Read the full file on GitHub · 572 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. yesterday First seen · 572 lines · 86 tokens per session scan B d3c9fac00e49

Subscribe to this mod's changes

secure-code-review is a skill published in the GitHub repository zebbern/claude-code-guide (4,600 stars, last pushed yesterday), licensed MIT. It adds 86 tokens to every session and 4,347 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 3 findings (cloud metadata endpoint, makes network calls, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.