Security audit

mimo-tts-wav

Security checks across malware telemetry and agentic risk

Overview

This appears to be a real MiMo text-to-speech skill, but it needs review because it can upload voice samples for cloning and send audio messages through Feishu without strong consent or privacy safeguards.

Install only if you are comfortable sending TTS text, style prompts, and any voice-cloning samples to Xiaomi MiMo, and sending generated audio to Feishu when that feature is used. Use voice cloning only for voices you are authorized to reproduce, protect the MiMo and Feishu credentials, and confirm the Feishu recipient before any send.

SkillSpector

By NVIDIA

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: 88% confidence
Finding: The activation guidance includes broad phrases like '说出来', '语音回复', and general 'read aloud/send voice message' scenarios, which can cause the skill to trigger in ordinary conversation without sufficiently specific user intent. In a skill that can write files, invoke shell scripts, and optionally send audio externally, overbroad activation increases the risk of accidental execution and unintended data handling.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill supports voice cloning from user-provided audio samples and sending audio to Feishu, but it does not provide clear privacy warnings, consent requirements, retention guidance, or disclosure of third-party processing. This is dangerous because biometric-like voice data and synthesized outputs may be uploaded, stored, or shared without the user fully understanding the privacy implications or having consent from the cloned voice owner.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: This script uploads user-provided audio content to Feishu and sends it as a message, but there is no explicit consent check, destination confirmation, or warning at execution time. In an agent context, this can cause unintended disclosure of sensitive spoken content to external recipients, especially if triggered automatically by a higher-level workflow.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The script base64-encodes a user-provided voice sample and sends it to a remote API for voice cloning, but it provides no explicit notice, consent prompt, or privacy warning at the point of transmission. Because voice prints are sensitive biometric data and the skill is specifically for voice cloning, silent transfer to a third party increases privacy, compliance, and misuse risk.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The script sends user-supplied synthesis text and voice-design context directly to a third-party remote API, and there is no in-code notice, consent gate, or data-minimization step before transmission. In a TTS skill, these fields may contain sensitive personal content, voice/persona descriptions, or confidential text, so undisclosed outbound transmission creates a real privacy and compliance risk.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.