Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent Chinese-language book recommendation and reading-list helper with modest local storage behavior and no evidence of hidden execution, exfiltration, or privilege abuse.
Install only if you are comfortable using a Chinese-language book helper that may store book lists or notes locally. Avoid putting sensitive personal reflections into notes unless you understand where the JSON files are stored and how to delete them.
The natural-language instructions, title, and usage examples are written in Chinese, which effectively forces a specific language for interaction. The file does not state that the skill is region-specific or offer users an alternative language or opt-in choice.
The usage section lists short natural-language phrases like “推荐小说” and “记录读书笔记” as activation examples, but it does not define whether these are exact triggers, examples only, or what requests should not invoke the skill. Because these phrases overlap with common everyday requests about books, the activation scope is ambiguous and could cause unintended invocation.
The trigger phrases are broad and map to very common book-related requests, which can cause the skill to activate in situations where the user did not specifically intend to use it. Because the skill has network access and local file write capability, overbroad invocation increases the chance of unintended data handling such as querying external services or storing reading notes without clear user intent.
The skill explicitly stores reading notes, ratings, and reading-status lists on disk, but it does not clearly warn users before collecting and persisting that personal data. This creates a privacy and consent risk, especially since notes may contain sensitive reflections or behavioral information and the storage path is documented without any mention of retention, protection, or opt-in consent.
The file’s natural-language description and all example interactions are written exclusively in Chinese, and the main command-matching logic depends on Chinese trigger phrases such as “推荐”, “添加”, and “书架”. This effectively constrains the skill to a specific language/locale without any opt-in or documented justification, which matches the language/locale policy violation criteria.
The help text and feature list present reading progress updates as a working capability, but this branch only parses a percentage and returns a success message without calling update_reading_progress or writing any data. This is an active contradiction between the documented intent of the command and the implemented behavior.
The module description and default help text list reading notes as a supported function, yet the command path for adding notes returns '功能开发中' instead of invoking add_reading_note. This directly contradicts the documented claim that note-adding is part of the skill's functionality.
No suspicious patterns detected.
