Security audit

Agentvibes Skill

Security checks for vulnerabilities and agentic risk

Overview

This TTS skill appears intended for audio generation, but it asks for broad command and file authority while its routing and data-flow instructions are too generic.

Review this skill before installing. Use it only for TTS/audio generation in a dedicated workspace, confirm output directories before batch export, avoid sending sensitive scripts or customer text to online engines unless you accept that data transfer, and do not let its broad automation wording justify unrelated shell or file operations.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (7)

Context-Inappropriate Capability

Medium

Confidence: 92% confidence
Finding: The skill declares broad capabilities including exec, write, glob, and grep despite being presented as a TTS-focused skill. This expands the agent's operational scope beyond what users would reasonably expect and increases the risk that the skill could be invoked for unintended file-system or command-execution tasks.

Context-Inappropriate Capability

Low

Confidence: 85% confidence
Finding: The description says to use the skill for general efficiency, automation, batch processing, and workflow optimization, which is much broader than speech synthesis. Overbroad positioning can cause an agent to route unrelated tasks to a skill that has powerful execution capabilities, increasing misuse risk.

Intent-Code Divergence

Medium

Confidence: 89% confidence
Finding: The input/output contract describes a generic text-processing skill and a generic result JSON rather than audio-generation behavior. This mismatch can mislead orchestrators or users into supplying arbitrary content and accepting unsafe downstream actions under an inaccurate interface description.

Intent-Code Divergence

Low

Confidence: 81% confidence
Finding: The feature summary advertises generic file handling, API integration, and command execution rather than narrowly scoped TTS operations. This inconsistency normalizes broader agent behavior and can hide the real security boundary of the skill.

Vague Triggers

High

Confidence: 94% confidence
Finding: The activation condition is extremely broad and tied to general productivity improvement, automation, and workflow optimization. In the context of a skill with exec and write access, this can cause inappropriate invocation for unrelated tasks, materially increasing the chance of dangerous actions being performed under an overly permissive skill.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The documentation describes batch audio export and file creation without warning about where files will be written, overwrite behavior, or the risk of modifying existing data. This can lead to accidental data loss or unsafe writes when the agent operates in a shared workspace.

Missing User Warnings

Medium

Confidence: 78% confidence
Finding: The skill references API-enabled behavior and external engine use but does not clearly warn that user text or content may be transmitted to external services. For a TTS skill, inputs may contain sensitive scripts, customer prompts, or proprietary material, so lack of a privacy notice increases data exposure risk.

Static analysis

No suspicious patterns detected.