ClawHub
Back to skill

Security audit

Lucasyao1985/clawdchat-analysis

Security checks across malware telemetry and agentic risk

Overview

The skill appears to do what it says: browse Moltbook, summarize public discussions, and save local reports, with no artifact-backed sign of hidden exfiltration or unsafe execution.

Install only if you are comfortable with the skill using browser automation on Moltbook and writing reports, and possibly raw scraped data, under ~/myassistant/chat/moltbook-daily/. Review or clean that directory periodically, especially on shared or synced machines.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger phrase "今天AI们在讨论什么" is broad enough to overlap with normal user conversation, which can cause the skill to activate unintentionally. In a skill that performs browser automation, web scraping, and local report creation, accidental activation can lead to unintended network activity and filesystem writes without the user explicitly requesting those side effects.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The README advertises automated scraping of dozens of posts, deep comment collection, and saving reports to disk, but it does not warn users about network access, rate/ToS considerations, or local file creation. This omission increases the risk that users invoke the skill without informed consent about potentially impactful actions, especially in an agent environment with browser automation capabilities.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill explicitly stores generated reports to a local path and documents optional storage of raw scraped data, but it does not require informed user consent or warn that persisted files may contain scraped third-party content and user activity history. This creates a privacy and data-retention risk, especially on shared machines or synced home directories where saved reports and raw JSON may be exposed beyond the current session.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.