Security audit
Security checks for vulnerabilities and agentic risk
The skill is a straightforward directory file-count reporter, but it asks for local file write access that its documented workflow does not need.
Review this before installing. The core task only needs read access to a user-selected directory, so remove or reject filesystem.write unless the skill is updated to clearly document confirmed, user-chosen file output.
The manifest description, procedural instructions, triggers, and example dialogue are all written to operate in Chinese, with no indication that the user can choose another language. This creates a locale/language policy concern because the skill appears to mandate a specific language without explicit opt-in or justification.
The trigger phrases are broad enough to activate on common conversational requests such as file counting or directory analysis outside a tightly scoped context. Overbroad activation can cause the skill to engage unexpectedly and prompt filesystem access when the user may have intended a different action.
The skill requests filesystem.write permission even though its documented behavior only returns a Markdown report in chat and does not require writing files. This creates unnecessary privilege that could enable unintended file creation or modification if the skill implementation or downstream agent behavior diverges from the documentation.
No suspicious patterns detected.
