ssh-penetration-testing

• Outcome: Review
• Latest audit: May 29, 2026 at 9:33 PM UTC
• Version:
  0.1.0

Overview

This SSH testing skill is instruction-only and openly describes penetration-testing use, but it includes under-scoped credential attacks and persistence steps that could preserve unauthorized access.

Install or use this only for systems you own or have explicit written authorization to test. Define target ranges, allowed accounts, rate limits, approved tools, tunnel destinations, logging, cleanup, and stopping conditions before use. Do not let an agent add SSH keys, reuse discovered private keys, search histories for passwords, or run brute-force workflows unless those actions are specifically authorized and documented.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Context-Inappropriate Capability

Medium

• Confidence: 98% confidence
• Finding: The skill explicitly instructs the operator to establish persistence by appending an SSH key to authorized_keys after access is gained. That goes beyond assessment or auditing and enables continued unauthorized access, which materially increases harm and changes the activity from testing to post-compromise persistence.

Context-Inappropriate Capability

Medium

• Confidence: 95% confidence
• Finding: This section expands into broad post-exploitation tradecraft, including searching the filesystem for keys, enumerating users, inspecting shell history for credentials, and harvesting SSH-related artifacts. In the context of an SSH testing skill, this is dangerous because it facilitates credential theft and lateral movement rather than focused service assessment.

Missing User Warnings

High

• Confidence: 97% confidence
• Finding: The skill provides destructive or state-changing persistence instructions without a prominent warning that they alter authentication state and may create unauthorized backdoor access. Even in a penetration-testing context, such changes can persist beyond the engagement and expose the target to compromise.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The skill includes brute-force and password-spraying procedures, as well as automated remote access attempts, but does not place a clear upfront warning about lockouts, service disruption, detection, and legal authorization before those steps. That omission makes misuse easier and increases the chance of operational harm on real systems.

VirusTotal

65/65 vendors flagged this skill as clean.