ClawHub
Back to skill

Security audit

播客下载工具免费版

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a podcast downloader, but it has overly broad activation language and unsafe filename handling that could write or delete files outside the intended podcast folder.

Review before installing. Use this only for explicit Xiaoyuzhou episode downloads, set PODCAST_DIR to a dedicated empty folder, and avoid running the shown shell workflow unless filenames are sanitized to prevent path traversal or unexpected overwrites/deletions.

SkillSpector

Vulnerability Patterns
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The skill description includes very broad trigger terms unrelated or loosely related to podcast downloading, such as document generation, automation, conversion, and marketing. In an agent ecosystem, overly broad activation cues can cause unintended invocation, leading the skill to run exec/read/write-capable workflows in contexts the user did not intend.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The skill's scope statement claims coverage for a wide set of scenario keywords, including generic automation and workflow language, instead of narrowly defining when the skill should activate. This broad matching surface increases the chance that an agent invokes a tool-enabled skill in inappropriate contexts, which is risky because the skill advertises exec/write capabilities.

Tool Parameter Abuse

High
Category
Tool Misuse
Content
# ...
# 6. 删除 m4a(可选)
if [ "$KEEP_M4A" != "true" ]; then
    rm "$EPISODE_DIR/${TITLE}.m4a"
fi
# ...
# 7. 提取节目笔记
Confidence
94% confidence
Finding
The script constructs filesystem paths from untrusted remote metadata fields such as TITLE and SHOW_NAME, then performs file creation and deletion operations including rm on those paths. If a malicious or malformed title contains path separators, traversal sequences, shell-sensitive characters, or special filenames, the skill could overwrite or delete unintended files within the agent's execution context.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.