ClawHub
Back to skill

Security audit

Telegram Voice STT (Windows)

Security checks for vulnerabilities and agentic risk

Overview

This skill is a narrow Telegram voice transcription helper with disclosed Google Cloud STT use and no bundled executable payload.

Installers should be comfortable with Telegram voice audio or transcripts being processed through the configured Google Cloud Speech-to-Text flow. The publisher should narrow the trigger wording so the skill activates only for actual Telegram voice-note/transcript contexts, but the current artifact does not show malicious or review-worthy behavior.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill description uses broad trigger phrases such as mentions of 'voice note, speech to text, STT, audio to text, or whisper on Telegram,' which can cause the skill to activate during ordinary conversation about transcription rather than only when an actual Telegram voice message is present. This can misroute user requests, override more appropriate skills, and cause the assistant to treat unrelated text as already-transcribed audio input.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.