Security audit
Security checks for vulnerabilities and agentic risk
This is a Baidu search helper that sends user-directed searches to Baidu using a configured API key, with some privacy and documentation caveats but no evidence of hidden, destructive, or deceptive behavior.
Install only if you are comfortable sending search terms to Baidu/Qianfan and storing a Baidu API key in the environment. Do not search for secrets or sensitive personal/company data with this skill, and consider removing the request-body debug print before use in logged environments.
代码的主要功能确实是调用百度 AI 搜索 API 进行搜索，这与声明的核心目的部分一致。但声明中一个关键能力是“优先使用 API 模式，配额不足时自动切换到浏览器模式”，而实际代码完全没有浏览器模式实现，也没有检测配额不足后切换执行路径的逻辑。代码只会在 API 调用失败时报错退出。此外，虽然可通过 resource_type_filter 等参数一定程度支持不同资源类型，但代码显式绑定的是 web_search API，并未体现独立的浏览器搜索能力。因此描述对实际能力有实质性夸大，应判定为不匹配。
The skill documents use of environment variables and outbound network access, but it does not declare any explicit tool scope or permissions boundary. That weakens reviewability and least-privilege controls, making it easier for a consumer to invoke a skill that can transmit user input and secrets to external services without clear upfront authorization metadata.
The skill encourages sending search queries to Baidu's API and possibly opening Baidu in a browser, but it does not prominently warn users that their queries leave the local environment and are disclosed to a third party. This can expose sensitive prompts, research topics, credentials accidentally included in queries, or regulated data to an external provider without informed consent.
The description emphasizes Chinese search support and presents the skill as Chinese-oriented, but does not explicitly state that the skill is region-specific or offer users a language/locale choice. Under the policy, forcing a specific language or locale without opt-in can be a natural-language policy violation.
The script logs the fully parsed request body, which includes the user’s search query and potentially other sensitive search parameters, before sending them to an external API. In a skill context, stdout is often captured by orchestration systems, logs, or telemetry pipelines, so this creates an unnecessary secondary disclosure channel in addition to the intended network transmission.
Natural-language content in this skill includes Chinese comments such as '禁用代理' and '从环境变量获取 API Key', which imposes a specific language context without opt-in. Under the policy, forcing a specific language without user choice can be a locale-policy violation unless clearly justified.
No suspicious patterns detected.
