Security audit
Security checks for vulnerabilities and agentic risk
This skill is a Chinese task-planning helper that reads specific project planning documents and writes a task plan only as part of its stated workflow.
Install this if you want a Chinese workflow for turning feature specs into TDD-oriented task plans. Expect it to read the named files in specs/ and to create a Markdown task plan under specs/features after confirmation. Review the destination path before allowing generation if an existing plan could be overwritten.
The trigger description uses broad phrases like '等意图' without defining clear boundaries, which can cause the skill to activate for loosely related requests. Over-broad activation increases the chance the agent performs planning, file reads, or file writes in contexts the user did not explicitly intend.
该技能的名称和说明整体以中文固定呈现，且没有任何文字说明可根据用户偏好切换语言或接受其他语言输入/输出。按照语言/地区策略，若技能默认强制单一语言而未提供选择或明确合理性说明，属于自然语言策略风险。
The skill instructs the agent to read project files such as specs and project context, and to operate according to their contents, but does not require clear disclosure or consent before accessing local files. This can violate user expectations and expose sensitive project information through unnecessary file access.
The document generation flow includes saving output to a local path, but it does not require warning the user or confirming before writing. Unannounced file creation or overwrite can lead to integrity issues, accidental data loss, or workspace modification the user did not intend.
The entire template, including headings, labels, and instructional text, is written only in Chinese, which establishes a fixed language requirement for generated task plans. Under the policy, a skill should not force a specific language or locale unless it offers a choice or clearly documents a justified region-specific constraint.
No suspicious patterns detected.
