SZZG007 TalkTrack Telegram

• Outcome: Review
• Latest audit: May 29, 2026, 9:33 PM
• Version:
  1.0.0

Overview

This is a coherent Telegram sales-assistant skill, but it needs review because it can drive customer outreach and store or reuse customer conversations without clear approval, privacy, or retention controls.

Install only if you can tightly control Telegram and Feishu access. Use dedicated least-privilege credentials, require human review before outbound customer messages, avoid unsolicited follow-ups, disable or bound auto-learning/cloud sync unless explicitly needed, and define how customer conversation data is stored, anonymized, retained, deleted, and shared.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (5)

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill explicitly supports Telegram 'human-like' sales conversations and automation for customer outreach, but it does not clearly warn users that it may contact external parties or send follow-up messages on their behalf. In a messaging context, silent automation can lead to unauthorized outreach, spam complaints, and platform-policy violations, especially because the skill is framed as an operational sales tool rather than a draft-only assistant.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The documentation says the system remembers customer conversation history and tracks script effectiveness, but it does not disclose retention, privacy, or lawful-use constraints for customer communications. This creates a real risk of collecting and storing personal or business data without notice, minimization, or access controls.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill describes automatically learning from successful and failed conversations to optimize future messaging, which implies reuse of customer communications for model or content improvement without any warning or approval workflow. That is dangerous because customer messages may contain sensitive commercial or personal information that should not be repurposed by default.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The follow-up guidance instructs sending another message after two days of customer silence, but it does not require consent, review, or compliance checks. In a sales-messaging workflow, that materially increases the risk of spam, harassment complaints, and account enforcement by Telegram or other service operators.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The environment configuration lists Telegram and Feishu credentials and storage paths but omits secure-handling guidance, which can normalize unsafe practices such as embedding tokens in plaintext configs or insecure local directories. If mishandled, these credentials could allow unauthorized messaging, data access, or compromise of integrated systems.

VirusTotal

66/66 vendors flagged this skill as clean.