Security audit

Telegram Toolkit

Security checks for vulnerabilities and agentic risk

Overview

The skill is a Telegram bot helper, but it gives the agent broad read/write/command authority and bulk messaging guidance without tight scoping or clear user-control requirements.

Review this skill before installing if you want tight control over agent actions. It is not clearly malicious, but it can lead an agent toward broad automation, command execution, file access, and large Telegram broadcasts without enough built-in guardrails. Use it only with explicit user approval for sends, file writes, command execution, and persistent storage.

SkillSpector

Vulnerability Patterns

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

Findings (5)

Intent-Code Divergence

Medium

Confidence: 95% confidence
Finding: The documented input/output contract describes a generic content-processing interface that does not match the advertised Telegram bot toolkit behavior. This can cause an agent to invoke the skill with incorrect parameters or trust incorrect outputs, increasing the chance of unsafe file/command actions through the declared read/exec/write tools under false assumptions.

Intent-Code Divergence

Medium

Confidence: 90% confidence
Finding: The edition description says the skill should be used for system monitoring, log analysis, alerting, and deployment management, which materially conflicts with its stated Telegram toolkit purpose. In an agentic environment, contradictory scope statements can trigger the skill in unrelated operational contexts, exposing file, execution, or messaging capabilities where they were not intended.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The invocation description uses broad language such as improving efficiency, automation, batch processing, and workflow optimization, which is likely to over-trigger the skill. Because the skill advertises read/exec/write capabilities, overly broad routing increases the risk that an agent selects it for tasks beyond Telegram messaging, leading to unintended side effects.

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The edition description contains ambiguous trigger conditions tied to general monitoring and deployment workflows, which broadens activation far beyond a Telegram bot toolkit. In a tool-enabled agent, this ambiguity can cause misrouting into a skill with command and file access, raising the chance of unnecessary or unsafe actions.

Missing User Warnings

Medium

Confidence: 87% confidence
Finding: The skill describes bulk messaging and also declares file and command capabilities, but it does not provide prominent, user-facing warnings about side effects such as mass outbound communication, data access, or system modification. In context, this is more dangerous because the skill targets automation and enterprise workflows, where silent high-impact actions can affect many users or systems quickly.

Static analysis

No suspicious patterns detected.