ClawHub

Telegram Voice STT (Windows)

Security checks across malware telemetry and agentic risk

Overview

This skill is a small Telegram voice-note transcription helper that clearly depends on Google Cloud transcription, with no bundled executable code or hidden persistence.

Install only if you are comfortable with Telegram voice notes being sent through Google Cloud and related OpenClaw/Dialogflow infrastructure for transcription and response handling. Use a limited service account, avoid sensitive voice content unless your retention and consent requirements are met, and consider tightening the trigger wording before broad deployment.

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 (2)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger description is broad enough to activate on generic mentions like 'speech to text', 'audio to text', or 'whisper on Telegram', which can cause the skill to run when the user did not explicitly intend Telegram voice-note handling. In a messaging-integrated assistant, unintended invocation can route content into a cloud transcription/processing path and alter how user input is interpreted, creating privacy and reliability risks.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill states that inbound voice is transcribed using Google Cloud Speech-to-Text and credential-backed infrastructure, but it does not clearly warn users that their audio and transcript may be sent to external cloud services. This creates a meaningful privacy and consent issue, especially for sensitive voice content, because users may not realize their data leaves the local chat context for third-party processing.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal