code-audit

Reviews source code for security vulnerabilities using Semgrep, CodeQL, and manual verification workflows.

21|8|Updated Aug 3, 2026
One-click install
npx skills add https://github.com/chengzongcai/reverse-skill-backup --skill code-audit-chengzongcai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: code-audit
Source: https://github.com/chengzongcai/reverse-skill-backup/tree/main/skills/code-audit
Command: npx skills add https://github.com/chengzongcai/reverse-skill-backup --skill code-audit-chengzongcai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Manual source-code security review is slow and inconsistent, while raw SAST scanner output is full of false positives. This Skill provides a structured, authorized white-box audit workflow that combines automated scanning with mandatory human verification to produce actionable findings. ## Core Features & Use Cases - Threat Modeling First: Maps trust boundaries, high-value assets, and input entry points before scanning begins. - Multi-Tool SAST Coverage: Supports Semgrep, CodeQL, Bandit, gosec, and SpotBugs/FindSecBugs across Python, Go, Java, and other languages. - Manual Verification Discipline: Requires reachability and exploitability checks on every scanner hit, covering IDOR, injection, hardcoded secrets, and crypto misuse. - Use Case: Given access to an authorized repository, run Semgrep with OWASP rules, then manually validate each finding and produce a report with data flow, PoC, CWE mapping, and fix recommendations. ## Quick Start Audit this repository for security vulnerabilities using Semgrep and manually verify each finding with fix recommendations.

Frequently Asked Questions about code-audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a security audit on source code with Semgrep?▼

Run semgrep --config auto . for automatic rule detection, or use targeted packs like semgrep --config p/owasp-top-ten. Then manually verify each hit for reachability and exploitability before reporting it as a finding.

What SAST tools should I use for different programming languages?▼

Semgrep covers multiple languages with fast custom rules, CodeQL provides deep data-flow analysis, Bandit targets Python, gosec and staticcheck cover Go, and SpotBugs with FindSecBugs handles Java.

Can I use this for code I do not have authorization to audit?▼

No. The workflow requires explicit authorization and confirmed source or repository access before starting. Without source code, binary targets should be routed to a reverse engineering skill instead.

Why do SAST scanners produce false positives?▼

Scanners match patterns without understanding runtime reachability, sanitization, or framework protections. Every finding must be manually validated for data-flow reachability and actual exploitability before being reported.

What is the difference between code audit and supply chain security review?▼

Code audit focuses on first-party code logic such as injection, broken authorization, and crypto misuse. Supply chain security covers third-party dependencies, build pipelines, and package integrity, which is handled by a separate skill.