ClawHub
Back to skill

Security audit

Bili Collection Pipeline

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed media transcription pipeline that downloads videos and writes local transcripts, with some dependency and privacy cautions but no evidence of hidden or malicious behavior.

Install only if you are comfortable with a tool that contacts Bilibili/YouTube, invokes local media tools, downloads media, and writes transcript files locally. Run it in a dedicated workspace with enough disk space, review playlist URLs first, update or pin dependencies before use, and do not use optional external LLM cleanup unless transcript content is safe to send to that provider.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill advertises and orchestrates network access, shelling out to external binaries, and reading/writing files, but it does not declare corresponding permissions. That mismatch is dangerous because users and any enforcement layer cannot accurately understand or constrain what the skill will do, increasing the chance of unintended downloads, command execution, or disk writes.

Missing User Warnings

Low
Confidence
87% confidence
Finding
The usage guidance describes batch transcription but does not warn that it will download potentially large media files and write multiple transcript and progress artifacts to disk. This can surprise users, consume storage/bandwidth, and cause accidental handling of copyrighted or sensitive media in local files.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The optional LLM post-processing step suggests sending transcript content to third-party APIs without a privacy warning. Transcripts may contain personal, proprietary, or copyrighted content, so forwarding them to external providers can create confidentiality, compliance, and data-retention risks.

Unpinned Dependencies

Low
Category
Supply Chain
Content
faster-whisper>=1.1.0
requests>=2.28.0
Confidence
90% confidence
Finding
requests>=2.28.0

Known Vulnerable Dependency: requests==2.28.0 — 5 advisory(ies): CVE-2024-47081 (Requests vulnerable to .netrc credentials leak via malicious URLs); CVE-2024-35195 (Requests `Session` object does not verify requests after making first request wi); CVE-2026-25645 (Requests has Insecure Temp File Reuse in its extract_zipped_paths() utility func) +2 more

High
Category
Supply Chain
Confidence
98% confidence
Finding
requests==2.28.0

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.