cs-security

Stars15

Forks1

UpdatedMarch 29, 2026 at 02:13

Full security audit: detect local vulnerabilities, cross-file taint flows, CVEs in dependencies, and polyglot sink patterns. Complete security scan pipeline.

Installation

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

Source

3D-Tech-Solutions

3D-Tech-Solutions/code-scalpel

View GitHub Repository View Creator Repositories

Download

Related occupationsSOC

Based on SOC occupation classification

Information Security AnalystsComputer and Mathematical Occupations·SOC 15-1212

SKILL.md

readonly

name	cs-security
description	Full security audit: detect local vulnerabilities, cross-file taint flows, CVEs in dependencies, and polyglot sink patterns. Complete security scan pipeline.
allowed-tools	["mcp__codescalpel__security_scan","mcp__codescalpel__cross_file_security_scan","mcp__codescalpel__unified_sink_detect","mcp__codescalpel__scan_dependencies"]
preamble-tier	1

/cs-security — Full Security Audit

Run a complete security analysis: find SQL injection, XSS, command injection, path traversal, tainted data flows, and vulnerable dependencies.

Usage

/cs-security
/cs-security src/api/

The Full Pipeline

Local analysis — Detect SQL injection, XSS, command injection, path traversal within files
- Taint analysis traces user input → dangerous functions
- Confidence threshold: 70% minimum
Cross-file analysis — Find vulnerabilities that span module boundaries
- Taint flows across function calls
- Identifies where untrusted data enters your system
Dependency scan — Check for known CVEs in your libraries
- Uses OSV (Open Source Vulnerabilities) database
- Shows version constraints to fix
Polyglot detection — Find sinks across Python, JavaScript, TypeScript, Java
- Frontend-backend vulnerability patterns
- TypeScript type-safety issues at system boundaries

Vulnerability Types Detected

SQL Injection — Raw SQL with user input
XSS — Unescaped output to HTML/DOM
Command Injection — User input to shell commands
Path Traversal — User input to file paths without validation
Tainted Data Flows — Untrusted data reaching sensitive functions
Type Erosion — JavaScript/TypeScript type safety breakdowns

Interpreting Results

HIGH confidence — Almost certainly a real vulnerability, fix immediately
MEDIUM confidence — Likely a vulnerability, investigate and validate
LOW confidence — Possible false positive, review code context

What Happens Next

Read the vulnerability report
Use /cs-extract to see the vulnerable code
Use /cs-refactor to fix it safely
Re-run /cs-security to confirm the fix

Pro Tip

Run this on your whole project regularly:

/cs-security .

See CLAUDE.md for the complete security workflow.

More from this repository

same repository

cs-analyze

3D-Tech-Solutions/code-scalpel

Analyze code structure: parse functions, classes, imports, and complexity metrics. Use to understand a file or project without reading the source code directly.

2026-03-2915

cs-extract

3D-Tech-Solutions/code-scalpel

Extract a specific function, class, or method from a file using Code Scalpel. Use when you need to show a symbol without reading the entire file—saves 95% of context.

2026-03-2915

cs-map

3D-Tech-Solutions/code-scalpel

Map project architecture: discover modules, build call graphs, visualize structure, trace critical paths. Complete architecture overview with dependency flows.

2026-03-2915

cs-policy

3D-Tech-Solutions/code-scalpel

Verify code compliance with policies and standards: HIPAA, SOC2, PCI-DSS, custom style guides. Validate policy file integrity with cryptographic verification.

2026-03-2915

cs-refactor

3D-Tech-Solutions/code-scalpel

Safe refactoring workflow: find all usages, trace dependencies, simulate changes, generate baseline tests, then apply with backup. Behavior-preserving refactor.

2026-03-2915

cs-setup

3D-Tech-Solutions/code-scalpel

Install and verify the Code Scalpel MCP server. Checks if codescalpel is already installed in Claude Code, runs `claude mcp add` if needed, and confirms connection by fetching capabilities.

2026-03-2915

/cs-security — Full Security Audit

Run a complete security analysis: find SQL injection, XSS, command injection, path traversal, tainted data flows, and vulnerable dependencies.

Usage

/cs-security /cs-security src/api/

The Full Pipeline

Local analysis — Detect SQL injection, XSS, command injection, path traversal within files

Taint analysis traces user input → dangerous functions
Confidence threshold: 70% minimum

Cross-file analysis — Find vulnerabilities that span module boundaries

Taint flows across function calls
Identifies where untrusted data enters your system

Dependency scan — Check for known CVEs in your libraries

Uses OSV (Open Source Vulnerabilities) database
Shows version constraints to fix

Polyglot detection — Find sinks across Python, JavaScript, TypeScript, Java

Frontend-backend vulnerability patterns
TypeScript type-safety issues at system boundaries

Vulnerability Types Detected

SQL Injection — Raw SQL with user input

XSS — Unescaped output to HTML/DOM

Command Injection — User input to shell commands

Path Traversal — User input to file paths without validation

Tainted Data Flows — Untrusted data reaching sensitive functions

Type Erosion — JavaScript/TypeScript type safety breakdowns

Interpreting Results

HIGH confidence — Almost certainly a real vulnerability, fix immediately

MEDIUM confidence — Likely a vulnerability, investigate and validate

LOW confidence — Possible false positive, review code context

What Happens Next

Read the vulnerability report

Use /cs-extract to see the vulnerable code

Use /cs-refactor to fix it safely

Re-run /cs-security to confirm the fix

Pro Tip

Run this on your whole project regularly:

/cs-security .

See CLAUDE.md for the complete security workflow.