Security audit
Security checks for vulnerabilities and agentic risk
The skill is mainly a podcast downloader, but its routing text is much broader than its real purpose and could cause agents to use command and file-write tools in unrelated coding tasks.
Review before installing. Use this only for Xiaoyuzhou podcast downloads, set PODCAST_DIR to a directory you control, and do not rely on its broad coding/debugging/deployment routing text. The package also references files that are not included, so expect manual setup or missing-command failures.
The manifest describes a narrowly scoped podcast downloader, but the invocation text repurposes it for broad coding, debugging, and deployment tasks. This scope mismatch can cause an agent to select and run a skill with Bash/Write/Edit permissions in contexts unrelated to podcast downloading, increasing the chance of unintended command execution or file modification.
The trigger keywords are overly broad and ambiguous, including generic terms that can match unrelated user requests. In an agent ecosystem, this can cause over-selection of a skill that has Bash, Read, Write, and Edit capabilities, making accidental execution in the wrong context significantly more dangerous than a mere documentation error.
The invocation conditions claim the skill should be used for broad software-development tasks even though its real behavior is a specialized downloader. This materially increases the risk of agent misrouting: the skill may be invoked during unrelated requests and then perform network access, shell execution, and filesystem writes outside the user's expected workflow.
No suspicious patterns detected.
