ClawHub
Back to skill

Security audit

Tmeet Skill

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly coherent for managing Tencent Meetings, but it asks agents to auto-install a global CLI and includes sensitive authentication and log-upload flows that need careful review.

Install only if you are comfortable granting an agent access to your Tencent Meeting account and meeting data. Before use, confirm any global npm installation yourself, treat OAuth links as sensitive, avoid uploading diagnostic logs unless you have reviewed what they may contain, and require explicit confirmation for meeting changes, participant actions, logout, and recording permission requests.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (4)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill instructs the system to automatically run `npm install -g @tencentcloud/tmeet@latest` when `tmeet` is missing, which modifies the host environment and pulls code from the network without an explicit user confirmation step. Installing the `latest` global package also creates a supply-chain and change-control risk, since behavior can change over time and a compromised or malicious package version would execute with the agent's privileges.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The documentation explicitly instructs agents to extract and fully display the OAuth authorization URL but does not warn that the URL may contain sensitive state, PKCE-related context, or one-time login material. In an agent setting, displaying or logging the full URL can leak authentication artifacts to unintended viewers, chat transcripts, or telemetry systems.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The documentation exposes a `--upload` option that sends local diagnostic logs to a server but does not clearly warn that logs may contain sensitive information such as tokens, meeting metadata, user identifiers, or system details. In a CLI/agent context, users or downstream automation may invoke this flag routinely for troubleshooting without understanding the privacy and data-exposure implications.

Ssd 3

Medium
Confidence
94% confidence
Finding
Requiring exposure of the full OAuth authorization URL increases the chance that login state or one-time authorization data is disclosed through agent output capture, terminal history, logging, or copied transcripts. In the context of a CLI skill explicitly designed for agent-mediated use, this is more dangerous because the agent is told to surface the full artifact and may do so in environments not controlled by the end user.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.