ClawHub

Telegram CLI

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent Telegram CLI wrapper, but installing it gives an agent access to private Telegram content and the ability to send messages as the authenticated user.

Install only if you trust the @cyberdrk/tg package or have reviewed the source. Authenticate only the Telegram account you intend to expose, confirm recipient and message text before any send/reply, and avoid broad search or sync commands for sensitive chats unless you are comfortable with local copies of that data.

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
91% confidence
Finding
The skill advertises broad triggers like checking messages, searching chats, looking up contacts, and sending messages using everyday language, which increases the chance of unintended activation. In this context, accidental invocation is more dangerous than usual because the skill can access private Telegram data and perform account-affecting actions such as sending messages to users or groups.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill documentation describes reading inboxes, searching all chats, listing members/admins, and sending messages, but it does not warn that these are privacy-sensitive and account-affecting actions. Without explicit warnings and consent expectations, users or upstream agents may invoke the skill in ways that expose sensitive message content or cause unintended outbound communication from the authenticated Telegram account.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal