ClawHub
Back to skill

Security audit

Bilibili Helper Tool Free

Security checks across malware telemetry and agentic risk

Overview

This Bilibili content-writing skill appears non-malicious, but it requests shell execution despite saying its core function is pure Markdown text generation.

Review before installing. The content-generation behavior is understandable, but the skill should not need shell execution; prefer installing only if exec is removed or you can restrict it to Bilibili-specific, user-directed use.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (3)

Intent-Code Divergence

Medium
Confidence
94% confidence
Finding
The documentation states the free edition is a pure Markdown/text-generation skill that does not require command execution, yet the manifest grants the exec tool. This mismatch can mislead users and host agents into trusting a broader privilege set than the skill functionally needs, increasing the risk that later prompts or hidden behaviors invoke shell execution unexpectedly.

Context-Inappropriate Capability

Medium
Confidence
96% confidence
Finding
This skill is presented as a helper for generating titles, descriptions, and tags, but it declares command-execution capability in allowed-tools. Granting shell access to a content-generation skill violates least privilege and enlarges the attack surface, because a prompt-influenced or modified skill could use exec to run local commands unrelated to its stated purpose.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger condition covers broad generic requests like marketing copy, writing, title optimization, and content creation, which overlap with many normal user tasks. Overbroad activation makes the skill more likely to be auto-invoked in unintended contexts, and because the skill also has unnecessary exec privileges, mistaken invocation becomes more dangerous than it would be for a harmless text-only skill.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.