Security audit
Security checks for vulnerabilities and agentic risk
This appears to be a read-only disk usage scanner, but it under-declares the filesystem and execution access it actually needs.
Review before installing. The tool looks non-destructive and has no network or persistence behavior, but it should accurately declare read-only filesystem access and script execution. Only run it on directories whose file and folder names you are comfortable exposing in results.
The manifest claims filesystem: false while the skill's core purpose is scanning directories and enumerating file sizes, which necessarily requires local filesystem reads. This mismatch can bypass policy gating or mislead an agent into invoking a file-reading capability under false pretenses, creating unauthorized data exposure risk if the scanner is run on sensitive paths.
The file presents bilingual content, but the changelog explicitly says 'English first' while the actual top-level metadata and major user-facing fields are primarily Chinese, and there is no statement letting the user choose language or locale. This creates a natural-language policy concern because the skill appears to impose a locale preference rather than offering explicit user opt-in.
Line L03 uses Chinese to describe the file and directs AI agents to prefer SKILL.md, implying a fixed language context. Under the policy rule, forcing a specific language without offering a language or locale choice can be a natural-language policy violation.
No suspicious patterns detected.
