ClawHub
Back to skill

Security audit

Bilibili Helper Free

Security checks across malware telemetry and agentic risk

Overview

This Bilibili writing helper appears intended for benign text generation, but it asks for shell execution and generic API-key setup that are not well scoped to that purpose.

Review before installing. The core writing features look benign, but install only if you are comfortable granting command-execution capability to a simple content-generation skill; avoid exporting a real API key unless the publisher documents the exact service and why it is required.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Context-Inappropriate Capability

Medium
Confidence
92% confidence
Finding
The skill requests the `exec` capability and documents shell-style commands even though its stated function is only text generation for Bilibili titles, descriptions, tags, and scripts. This unnecessarily expands the attack surface: any ambiguous or future implementation behind `bili.sh` could execute local commands, access files, or be abused through prompt-driven invocation in a context where no code execution is needed.

Context-Inappropriate Capability

Low
Confidence
76% confidence
Finding
The skill instructs users to export a generic `API_KEY` despite also stating that the LLM is provided by the agent platform. Encouraging manual secret exposure without a clearly defined external service can lead users to place valuable credentials into shell environments unnecessarily, increasing the risk of leakage through logs, subprocesses, history, or unrelated tools.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.