Security audit

Telegram Chat Tool Free

Security checks across malware telemetry and agentic risk

Overview

This is a real Telegram bot setup skill, but it needs Review because it asks for group-admin/read-all-message access and bot-token handling without strong least-privilege and privacy controls.

Install only if you are comfortable giving a Telegram bot access to the selected groups. Limit the bot's admin permissions, keep privacy mode on unless full group-message monitoring is truly needed, get group participant consent where appropriate, store the bot token in a secret store or protected environment variable, and rotate the token if it was pasted into logs, chats, or repositories.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Description-Behavior Mismatch

Medium

Confidence: 84% confidence
Finding: The skill describes itself as a lightweight, limited free edition, yet its runtime guidance still encourages external network interaction and shell-assisted troubleshooting (for example using ping/curl via exec-capable agents). In an agent setting, that expands the operational scope beyond passive documentation and can lead to unintended network access or command execution without clearly bounded safety controls.

Intent-Code Divergence

High

Confidence: 80% confidence
Finding: The document gives inconsistent statements about whether the free version can proactively send messages. In agent workflows, contradictory capability claims can cause the model to overreach and perform message-sending actions the user did not expect, increasing the chance of unauthorized communications or policy bypass.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The trigger condition is so broad that it covers generic communication and notification use cases, making accidental activation more likely. In an agent environment with exec enabled, overbroad triggering can cause the skill to handle requests outside its intended Telegram-only scope and invoke sensitive actions unnecessarily.

Missing User Warnings

High

Confidence: 96% confidence
Finding: The skill instructs users to disable Telegram Bot privacy mode so the bot can read all group messages, but it does not prominently warn that this enables collection of every message in the group. That creates a meaningful privacy and compliance risk, especially in team or multi-user groups where participants may not expect full-message monitoring by an agent-controlled bot.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The skill includes bot token configuration examples without a strong, immediate warning about secure secret handling. In practice, users may paste live credentials into files, chats, logs, or repositories, which could let an attacker take over the Telegram bot and impersonate it or access its communications.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.