ClawHub
Back to skill

Security audit

快速部署飞书的多个Bot军团，只需要在飞书配置好机器人，然后就可以马上创建并联动

Security checks across malware telemetry and agentic risk

Overview

This skill appears purpose-aligned, but it needs review because it can store Feishu secrets, change live OpenClaw bot routing, and restart the gateway without a clear confirmation or dry-run step.

Install only if you intend to let the skill modify your OpenClaw agent, Feishu channel, binding, and gateway configuration. Before running it, ask for a dry run and approve the exact changes; use least-privilege Feishu credentials; avoid pasting production appSecret into general chat; and confirm how to remove the created agent, account, binding, and stored secret afterward.

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
89% confidence
Finding
The skill advertises automatic credential insertion, route binding, and gateway restart, but the top-level description does not make the operational impact explicit enough for a user to understand that live configuration will be modified immediately. In an infrastructure-management skill, hidden or under-emphasized state-changing behavior increases the risk of accidental production changes and service disruption.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The quick-start block asks users to paste appSecret directly into the skill workflow without a strong privacy warning, despite the skill subsequently writing that secret into configuration and using commands that may expose it through logs or config reads. Collecting sensitive credentials in a casual invocation format materially raises the chance of secret leakage via chat history, audit logs, or operator mistakes.

Ssd 3

Medium
Confidence
92% confidence
Finding
The troubleshooting guidance includes commands such as reading Feishu account configuration, which can surface stored credentials or encourage operators to inspect secret-bearing configuration directly. Even if intended for debugging, normalizing direct reads of secret material increases exposure in terminal history, logs, screenshots, and shared support sessions.

Ssd 3

Medium
Confidence
96% confidence
Finding
The core workflow explicitly writes user-supplied appId and appSecret into persistent configuration in plaintext-accessible form, making those credentials retrievable by later commands or anyone with filesystem/config access. In the context of bot deployment automation, this is especially risky because the skill is designed to operationalize production credentials and then continue running validation and admin commands around them.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.