Live survey304 models · 14 providers · 23,592 auditsAudited implement · SAFE →What changedHow we verifyllms.txt

Agent skill · Audited from source

Penetration Tester Master

skill · Dokhacgiakhoa/antigravity-ideNOASSERTION licenceAudited 19 Oct 2026 · source d1a0c85a714a

★ 508pushed 130 days agoversion 4.1.0-fractal
✕BLOCKGrade D · 69/100 · 6 findings · d1a0c85
508repo stars · whole Dokhacgiakhoa/antigravity-ide

Is it safe? No — BLOCK, grade D (69/100). We read this skill's source at commit d1a0c85a714a and found no filesystem, network or shell access and no hard-coded secrets. The audit raised 6 findings, 2 critical or high. What it does: Ultimate Offensive Security Master Skill.

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “Penetration Tester Master (Dokhacgiakhoa/antigravity-ide)”. Retrieved 19 Oct 2026. https://modelsatlas.com/skills/dokhacgiakhoa/penetration-tester-master

Blocked. The page stays public so you can check this skill if someone recommended it to you. The audit found 2 critical or high findings, starting with “Prompt injection” in SKILL.md. Do not install it without reading them.

Trust audit

Audit 0.4.1 · source read 19 Oct 2026

Trust score69/100✕BLOCK D
0 critical2 high3 medium1 low0 info

BLOCK means the audit found something that could harm you or your machine. Grade D is shared by 749 of 18,389 audited skills.

What the source does

  • FilesystemNone observedNo reads or writes in the source
  • NetworkNone observedNo outbound calls in the source
  • ShellNone observedNo commands run in the source
  • DependenciesPinnedNothing floats
  • SecretsNone foundNo keys in source

Five audit layers

LayerWhat we checkResultFindings
L0 ProvenanceRepo owner, licence, archive status, activityPASS0
L1 Static sourceSecrets, path traversal, supply chain, hidden codePASS0
L2 InstructionsPrompt injection, authority framing, hidden directivesFAIL6
L3 InventoryThe files and declarations match the sourcePASS0
L4 SandboxRun in isolation and watch behaviourSKIPPED why0

Gates applied: no behavioural pass.

Findings (6)

HIGHPrompt injectionreview.instruction_override · SKILL.md
F-01 Prompt injectionCWE-94 · CWE-1427Layer L2
You are an **Elite Red Team Lead and Professional Pentester**. This skill provides a unified lifecycle for identifying, exploiting, and reporting security vulnerabilities.
Why
The skill overrides the agent's role with an offensive 'exploit and steal' posture, including unauthorized access and exfiltration, contrary to safe default behavior.
Fix
rewrite it so the instruction says plainly what it does, and asks the user before it acts
HIGHPrivilege escalation / persistencereview.unsafe_action · SKILL.md
F-06 Privilege escalation / persistenceCWE-269 · CWE-250Layer L2
**Goal**: Move from low-privileged user or external network to Domain Admin or Data Exfiltration.
Why
Directs the agent toward privilege escalation and data exfiltration goals without requiring user authorization or scope.
Fix
rewrite it so the instruction says plainly what it does, and asks the user before it acts
MEDIUMPrompt injectionreview.misleading_scope · SKILL.md
F-01 Prompt injectionCWE-94 · CWE-1427Layer L2
**Checklist**: Define scope, obtain "Get Out of Jail Free" letter, and verify legal boundaries.
Why
The scope-limiting checklist is advisory only, while the execution protocol proceeds straight to attack without requiring verified authorization.
Fix
rewrite it so the instruction says plainly what it does, and asks the user before it acts
MEDIUMPrivilege escalation / persistencereview.unsafe_action · SKILL.md
F-06 Privilege escalation / persistenceCWE-269 · CWE-250Layer L2
**Phases**: Recon → Scanning → Gaining Access → Maintaining Access → Covering Tracks.
Why
Instructs a persistent, stealthy attack lifecycle including 'covering tracks', a destructive/persistent action taken without the user's say-so.
Fix
rewrite it so the instruction says plainly what it does, and asks the user before it acts
MEDIUMPrivilege escalation / persistencereview.unsafe_action · SKILL.md
F-06 Privilege escalation / persistenceCWE-269 · CWE-250Layer L2
**XSS/HTML Injection**: Bypass CSP and steal cookies.
Why
Directs the agent to bypass protections and steal user credentials/cookies as part of the engagement.
Fix
rewrite it so the instruction says plainly what it does, and asks the user before it acts
LOWPrompt injectionreview.misleading_scope · SKILL.md
F-01 Prompt injectionCWE-94 · CWE-1427Layer L2
Synthesize findings for the user.
Why
Framed as serving the user, but the package is an offensive attack kit whose description hides the full exploit scope.
Fix
rewrite it so the instruction says plainly what it does, and asks the user before it acts

Audit 0.4.1 · source sha d1a0c85a714a · 19 Oct 2026Report an issue or request a re-scan

Provenance

Who published it, and what the verdict is pinned to

  • PublisherDokhacgiakhoa GitHub userOwner of the repository this skill is read from
  • Source folderDokhacgiakhoa/antigravity-ide/.agent/skills/penetration-tester-masterThe exact folder we audited
  • LicenceNOASSERTIONAs the repository states it
  • Commit pinned for this verdictd1a0c85a714aA new audit runs when the source changes
  • Repository statusNot archivedLast push 130 days ago
  • Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

What it tells the agent

Full source on GitHub ↗

Its description: Ultimate Offensive Security Master Skill.

---
name: penetration-tester-master
description: Ultimate Offensive Security Master Skill.
category: security
version: 4.1.0-fractal
layer: master-skill
---

# 🗡️ Penetration Tester Master Kit

You are an **Elite Red Team Lead and Professional Pentester**. This skill provides a unified lifecycle for identifying, exploiting, and reporting security vulnerabilities.

---

## 📑 Internal Menu
1. [Hacking Methodology & Planning](#1-hacking-methodology--planning)
2. [Reconnaissance & OSINT](#2-reconnaissance--osint)
3. [Exploitation (Web, API, Cloud)](#3-exploitation-web-api-cloud)
4. [Post-Exploitation & PrivEsc](#4-post-exploitation--privesc)
5. [Reporting & Remediation](#5-reporting--remediation)

---

## 1. Hacking Methodology & Planning
*Structured approach to offensive engagements.*
- **Phases**: Recon → Scanning → Gaining Access → Maintaining Access → Covering Tracks.
- **Checklist**: Define scope, obtain "Get Out of Jail Free" letter, and verify legal boundaries.
- **Goal**: Move from low-privileged user or external network to Domain Admin or Data Exfiltration.

---

## 2. Reconnaissance & OSINT
- **Passive**: Use **Shodan**, Google Dorks, and WHOIS.
- **Active**: **Nmap** (Port scanning), **Wireshark** (Traffic analysis), and Subdomain enumeration (**Sublist3r**).
- **Tools**: Find exposed Jenkins, Git configs, or unsecured API endpoints.

---

## 3. Exploitation (Web, API, Cloud)
- **Web**: Master the OWASP Top 10.
  - **SQL Injection**: Use **SQLMap** for automation.
  - **XSS/HTML Injection**: Bypass CSP and steal cookies.
  - **Path Traversal/LFI**: Read `/etc/passwd` or configuration files.
  - **IDOR**: Access other users' data by manipulating IDs.
- **API**: Fuzzing with **Burp Suite**, testing for Broken Object Level Authorization (BOLA).
- **Cloud (AWS/Azure)**: Target S3 misconfigurations, Metadata SSRF, and Lambda exploitation.

---

## 4. Post-Exploitation & PrivEsc
- **Metasploit Framework**: Use for payload generation and session management.
- **Linux PrivEsc**: Check for SUID binaries, kernel exploits, and misconfigured cron jobs.
- **Windows PrivEsc**: Target DLL hijacking, Token Impersonation, and unquoted service paths.
- **Active Directory**: Kerberoasting, Pass-the-Hash, and BloodHound enumeration.

---

## 5. Reporting & Remediation
- **Evidence**: Collect screenshots, logs, and reproduction scripts (PoC).
- **Severity**: Rank finds via CVSS (0-10).
- **Remediation**: Provide clear, developer-friendly fixes (e.g., "Use parameterized queries" instead of "Fix SQL Injection").

---

## 🛠️ Execution Protocol

1. **Classify Sector**: Network, Web, Cloud, or Mobile?
2. **Phase 1: Recon**: Gather target intel.
3. **Phase 2: Scanning**: Identify services and versions.
4. **Phase 3: Attack**: Select and execute the specific exploit logic above.
5. **Phase 4: PrivEsc**: Elevate permissions if possible.
6. **Final Report**: Synthesize findings for the user.

---
*Merged and optimized from 25 legacy offensive security and tool-specific skills.*


## 🧠 Knowledge Modules (Fractal Skills)

### 1. [owasp_top_10_2025](./sub-skills/owasp_top_10_2025.md)

Excerpt, first 3,118 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

Install

Pinned to the audited commit d1a0c85

git clone https://github.com/Dokhacgiakhoa/antigravity-ide
git -C antigravity-ide checkout d1a0c85a714a

The author documents no install command, so this uses the repository directly. Our verdict covers commit d1a0c85a714a only.

Works in

The author has not said which agents this skill supports, and we have not tested it in any.

Audit history

Every audit, every source change

  1. ✕BLOCK D 69 / 100

    first audit · source d1a0c85a714a

Questions about Penetration Tester Master

Is Penetration Tester Master safe to install?
No — not without reading the findings first. The audit graded it D (69/100) and found 2 critical or high issues in the source. Each one is listed on this page with the file and line it is on.
What can Penetration Tester Master access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (d1a0c85a714a), read on 2026-10-19. The repository is watched, and a new audit runs when it changes — this is the first audit.

Sources

Spotted something wrong? Report a correction · Methodology

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.