Security audit
Security checks for vulnerabilities and agentic risk
This looks like a disclosed video parsing/download helper, but users should be careful with its optional local HTTP service and the missing referenced binaries.
Install only if you intentionally want a Chinese-language video parsing/de-watermark helper and trust any platform binaries supplied with the real package. If using service mode, keep it bound to localhost, do not expose it publicly without authentication, and treat parsed media links as temporary third-party network data.
The README claims the tool is 'only local' and 'safe/reliable' while also documenting an HTTP service mode. That mismatch can mislead users about the attack surface and trust assumptions, causing them to expose a listener or process remote requests without realizing the security implications.
The documentation encourages automatic link parsing, downloading, and running an HTTP service but does not explain network exposure, logging, temporary link handling, or what data may be sent or received. In a skill that processes untrusted URLs and can open a server, missing safety guidance increases the risk of accidental exposure, unsafe deployment, or misuse with sensitive content.
The skill instructs the agent/user to run shell scripts and local binaries, but it declares no explicit tool scope or permissions boundary. That creates an authorization gap where shell-capable behavior can be invoked without a clear least-privilege contract, increasing the chance of unintended command execution in environments that auto-enable skills.
The trigger list is very broad, including generic phrases like '下载视频', 'parse video', and platform names, which can cause the skill to activate for ordinary user requests not specifically asking to run this skill. Over-broad activation expands the attack surface by making it easier for risky shell actions or binary execution to be invoked unexpectedly.
This shell script includes its description, usage instructions, examples, and runtime messages entirely in Chinese. For a general-purpose skill, forcing a specific language without opt-in is a natural-language policy issue because users are not given a locale choice and no region-specific justification is stated.
All visible instructions and usage examples are presented only in Chinese, which can impose a language requirement on users without giving them a choice. The policy specifically flags language or locale constraints when they are forced without opt-in or documented justification.
The documentation tells users to start an HTTP service and exposes a localhost Web UI, but it does not clearly warn about the security implications of opening a listening port. Even if bound locally, users may expose the service unintentionally via port forwarding, shared environments, or modified defaults, increasing the risk of unauthorized access.
The description is written only in Chinese ("跨平台视频去水印解析工具，支持20+平台") with no indication that other languages are supported or that the user can choose a locale. This can violate language/locale policy when a skill presents a fixed language without opt-in or documented regional justification.
No suspicious patterns detected.
