Claude Skill

audit-security

Is this safe?

LLM Mart · 0 points · 2 views 0 listing impressions 0 install-command copies
Download dhaupin-vant-models_public_vant_skills_vant-skill-audit-security-e1e348e.zip · 1 KB

Install

skills CLI npx skills add https://github.com/dhaupin/vant/tree/main/models/public/vant/skills/vant-skill-audit-security
Claude Code claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install dhaupin-vant@llmmart
Git git clone https://github.com/dhaupin/vant.git

The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole dhaupin/vant collection as a plugin from our marketplace. Git is the plain clone.

Skill manifest

Security Audit

Is this safe?


When To Run

  • Before deploy
  • After adding auth/credentials
  • New endpoints
  • External inputs

What To Check

1. Credentials

# Hardcoded secrets?
grep -rn "sk-\|api_key\|password\|secret" . --include="*.js"
grep -rn "Bearer\|Token" . --include="*.js"
Check Issue Fix
Hardcoded API key HIGH Use env var
Password in code HIGH Use env
Token in logs MEDIUM Redact

2. Inputs

// Never trust user input
req.params.id
req.body.data
query.string
Check Issue Fix
SQL from input HIGH Parameterize
Eval input HIGH Remove eval
Shell from input HIGH No shell
File from input HIGH Validate path

3. Auth

// Check auth exists
function protected() {
  if (!req.user) return error
  // Good
}
Check Issue Fix
No auth check HIGH Add auth
Auth bypassed HIGH Fix
Weak auth MEDIUM Strengthen

4. Network

Check Issue
HTTP not HTTPS Use HTTPS
Credentials in URL Headers only
No CORS Set CORS

5. Filesystem

// Check file access
fs.readFile(userPath)
Check Issue Fix
Path traversal HIGH Sanitize
Read any file HIGH Validate
Write anywhere HIGH Restrict

Output

## Security Audit - [file]

### Credentials
- [PASS/FAIL] Hardcoded keys: [details]

### Inputs  
- [PASS/FAIL] SQL injection: [details]

### Auth
- [PASS/FAIL] Protected: [details]

### Summary
| Severity | Count |
|----------|-------|
| HIGH     | 0     |
| MEDIUM   | 0     |
| LOW     | 0     |

Severity

Severity Meaning
HIGH Exploit - fix now
MEDIUM Risk - fix soon
LOW Note - fix Optional

Role: Security Auditor Input: Code to review Output: Issues found

Files (vant)
  • SKILL.md 2.1 KB
    ---
    name: audit-security
    description: Is this safe?
    license: MIT
    metadata:
      author: vant
      version: "1.0"
    ---
    
    # Security Audit
    
    > Is this safe?
    
    ---
    
    ## When To Run
    
    - Before deploy
    - After adding auth/credentials
    - New endpoints
    - External inputs
    
    ---
    
    ## What To Check
    
    ### 1. Credentials
    
    ```bash
    # Hardcoded secrets?
    grep -rn "sk-\|api_key\|password\|secret" . --include="*.js"
    grep -rn "Bearer\|Token" . --include="*.js"
    ```
    
    | Check | Issue | Fix |
    |-------|-------|-----|
    | Hardcoded API key | HIGH | Use env var |
    | Password in code | HIGH | Use env |
    | Token in logs | MEDIUM | Redact |
    
    ### 2. Inputs
    
    ```javascript
    // Never trust user input
    req.params.id
    req.body.data
    query.string
    ```
    
    | Check | Issue | Fix |
    |-------|-------|-----|
    | SQL from input | HIGH | Parameterize |
    | Eval input | HIGH | Remove eval |
    | Shell from input | HIGH | No shell |
    | File from input | HIGH | Validate path |
    
    ### 3. Auth
    
    ```javascript
    // Check auth exists
    function protected() {
      if (!req.user) return error
      // Good
    }
    ```
    
    | Check | Issue | Fix |
    |-------|-------|-----|
    | No auth check | HIGH | Add auth |
    | Auth bypassed | HIGH | Fix |
    | Weak auth | MEDIUM | Strengthen |
    
    ### 4. Network
    
    | Check | Issue |
    |-------|-------|
    | HTTP not HTTPS | Use HTTPS |
    | Credentials in URL | Headers only |
    | No CORS | Set CORS |
    
    ### 5. Filesystem
    
    ```javascript
    // Check file access
    fs.readFile(userPath)
    ```
    
    | Check | Issue | Fix |
    |-------|-------|-----|
    | Path traversal | HIGH | Sanitize |
    | Read any file | HIGH | Validate |
    | Write anywhere | HIGH | Restrict |
    
    ---
    
    ## Output
    
    ```
    ## Security Audit - [file]
    
    ### Credentials
    - [PASS/FAIL] Hardcoded keys: [details]
    
    ### Inputs  
    - [PASS/FAIL] SQL injection: [details]
    
    ### Auth
    - [PASS/FAIL] Protected: [details]
    
    ### Summary
    | Severity | Count |
    |----------|-------|
    | HIGH     | 0     |
    | MEDIUM   | 0     |
    | LOW     | 0     |
    ```
    
    ---
    
    ## Severity
    
    | Severity | Meaning |
    |----------|---------|
    | HIGH | Exploit - fix now |
    | MEDIUM | Risk - fix soon |
    | LOW | Note - fix Optional |
    
    ---
    
    **Role**: Security Auditor **Input**: Code to review **Output**: Issues found
    

Comments (0)

Sign in to join the conversation.

No comments yet.

Reviews (0)

No reviews yet.