Security audit

Ai Coding Tools Full Suite

Security checks for vulnerabilities and agentic risk

Overview

This is a reference skill for comparing AI coding tools, with broad routing and manual helper scripts but no evidence of hidden execution, credential access, or data exfiltration.

Install this as a documentation and reference skill. Use explicit commands when possible to avoid accidental routing, and do not treat the analyzed file-write, shell, browser, or context-persistence capabilities as permission for your agent to use those tools unless you separately approve them.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The skill allows activation via very broad natural-language input, including essentially any direct description of need. Overly permissive triggering can cause unintended invocation, context hijacking, or accidental routing when users are discussing related topics rather than explicitly requesting this skill.

Vague Triggers

Medium

Confidence: 94% confidence
Finding: The keyword routing table uses generic phrases such as '学习', '入门', '教程', '概览', and other common terms without contextual constraints. This increases the chance that unrelated user messages will be captured and routed into the skill, which can lead to prompt-scope confusion, misfires, or unintended override of a user's actual request.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The summary explicitly states that a one-sentence description of user need is enough to invoke the skill. This reinforces an activation model that is too broad and may cause the skill to engage without a deliberate user opt-in, increasing the risk of unintended prompt capture and workflow interference.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The skill registers very broad triggers such as '/ai-tools' and '/aitools', which can plausibly match generic user requests about AI tools rather than an explicit request for this specific skill. This increases the chance of unintended invocation, causing the skill to intercept conversations outside its narrow scope and potentially steer users into its bundled content without clear consent.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: This document explicitly describes powerful capabilities including file writes, shell command execution, web fetching, and collection of system/user/project context, but provides no user-facing safety boundaries, consent guidance, or warnings about privacy and system impact. In a skill package meant to guide AI coding tools and agent design, that omission can normalize unsafe use of high-risk actions and increase the chance of unintended data exposure, destructive commands, or over-collection of environment information.

Static analysis

No suspicious patterns detected.