Security audit

CTF Router

Security checks for vulnerabilities and agentic risk

Overview

This skill is a CTF challenge routing and setup helper whose local checks and tool suggestions are disclosed and aligned with CTF use.

Install this only if you want CTF-oriented routing and setup help. Run the env-check script only when you are comfortable with it printing local Python, tool, WSL, and common install-path information to your session; review competition rules before using AI assistance or offensive-security tools.

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (5)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 84% confidence
Finding: The skill advertises and operationalizes file reading, shell command usage, and potential network-enabled tooling without any declared permission boundaries. In an agent environment, this creates an authorization gap: the orchestrator and user may believe the skill is only doing lightweight routing, while it can direct execution of local inspection commands and external tools that touch the filesystem, shell, and possibly remote targets.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 94% confidence
Finding: The documented purpose is challenge triage, but the content also instructs environment auditing, package enumeration, tool discovery, and subprocess-based WSL interrogation. That mismatch is dangerous because it hides materially broader system-interaction behavior inside a skill that users would reasonably expect to be classification-only, increasing the chance of unreviewed host reconnaissance and command execution.

Description-Behavior Mismatch

Medium

Confidence: 89% confidence
Finding: This skill is described as a CTF challenge triage router, but the bundled script audits the host environment, enumerates installed tools, and prints installation guidance for offensive/security utilities. That scope expansion is dangerous because it conditions a routing-only skill to collect local reconnaissance data and facilitate capability buildup unrelated to simple challenge classification.

Context-Inappropriate Capability

Medium

Confidence: 92% confidence
Finding: The code performs broad host reconnaissance by probing PATH, common install directories, WSL distributions, interpreter details, architecture, and numerous security tools. In the context of a triage-only router, this is unjustified access to environmental information and increases risk because an adversarial or overprivileged skill could use the same pattern to fingerprint the analyst’s or user’s machine.

Vague Triggers

Medium

Confidence: 81% confidence
Finding: The trigger language is extremely broad: 'any ctf, solving, get flag' style request should invoke this skill first. Overbroad activation can cause the skill to intercept unrelated or ambiguously phrased requests, unnecessarily exposing the agent to shell/tooling guidance and potentially routing benign conversations into offensive-security workflows.

Static analysis

No suspicious patterns detected.