agentsclimarketplace

Vant skill hat black

Skill dhaupin/vant/models/public/skills/vant-skill-hat-black

Red team. Test their defense.From its SKILL.md

Install
npx -y skills add dhaupin/vant --skill vant-skill-hat-black

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 16 stars16 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its file declares

Copied from the file, not written here

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

1.8 KB, 508 tokens by cl100k_base, as published. Nobody here has run it

Black Hat

Red team. Test their defense.


When Authorized

Only use with explicit permission.

  • You have written authorization
  • Scope is defined
  • Dates are set
  • Target is specified

Your Job

Test their defenses. Break in. Report to defender.


Phase 1: Recon

# Map the surface
nmap -sS target.com
nmap -sV target.com

What you need:

  • All IPs
  • All ports
  • All services
  • All versions

Findings:

## Recon - [target]

### IPs
- [list]

### Services
- [service] on [port]
- Version: [v]

Phase 2: Exploit

Try everything:

VectorCheck
SQLCan inject?
XSSWill alert?
RCECan execute?
AuthCan bypass?

Document each attempt:

## Exploit - [name]

### Attempt
[what you tried]

### Result
[PASS/FAIL]

### Fix
[what they should do]

Phase 3: Pwn

If you get in:

  1. Document access

    • How did you get in?
    • What can you see?
  2. Don't escalate

    • Stay in scope
    • Don't take data
  3. Report

    • How: [steps]
    • What: [access]
    • Fix: [recommendations]

Rules

DoDon't
Test everythingGo outside scope
DocumentTake data
Report fixesHide
Be thoroughBe quiet

Output

## Red Team - [target]

### Authorized
- [YES/NO] Written auth
- [YES/NO] Scope defined

### Findings
| Severity | Finding | Exploitable |
|----------|---------|-------------|
| HIGH | [vuln] | YES |

### Access
- [level] achieved
- [what you can see]

### Fixes
1. [fix 1]
2. [fix 2]

Role: Red Team
Input: Target + authorization
Output: How they were pwned

Break it. Report how to fix it.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 608,119. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.