Security audit
Security checks for vulnerabilities and agentic risk
This looks like a document-analysis skill, but it asks for command execution and file-writing powers that are not explained.
Review this before installing. The skill does not show malware-like behavior, but it requests command execution and file-writing authority for what is described as document analysis. Prefer installing only if the publisher narrows the tool permissions or clearly documents when commands and file changes are allowed, with explicit user confirmation.
The display name, summaries, and the full skill content are presented in Chinese, with no indication that users may choose another language or that the skill is intentionally restricted to a Chinese-language or region-specific context. That creates a natural-language policy concern because it imposes a locale/language choice without opt-in.
The description is broadly phrased as applicable to generic development scenarios and reusable workflows, without clear trigger boundaries or narrow activation criteria. Overly broad activation language can cause the agent to invoke this skill in contexts unrelated to long-document analysis, exposing exec/write capabilities more often and making prompt-trigger confusion easier.
The tool list exposes read, exec, and write, but the user-facing markdown does not prominently warn that the skill can execute commands and modify data. This is dangerous because users may reasonably treat it as a passive analysis skill while it actually has side-effecting capabilities, undermining informed consent and increasing the risk of covert file changes or command execution.
The skill advertises itself as a document reasoning/analysis tool, yet declares both exec and write capabilities without any task-bound justification, guardrails, or constrained execution model. This creates an unnecessary privilege expansion: a user invoking a seemingly analytical skill could cause filesystem changes or command execution far beyond the documented purpose, increasing the blast radius of prompt injection or accidental misuse.
