Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed Kuaishou research skill that uses a third-party API token to fetch public video, profile, and comment data, then saves results locally.
Install only if you are comfortable sending Kuaishou keywords, profile/video URLs, and requested result parameters to guaikei.com using your GUAIKEI_API_TOKEN. Review or clear the generated logs directory if the resulting research data should not remain on disk.
声明描述的是一个面向快手竞品研究/内容调研的业务技能，核心能力应涉及快手平台数据查询、内容对比或分析。给出的代码块仅实现了通用 CLI 参数解析与帮助文本生成，包括读取参数值、校验类型、处理布尔值、位置参数和输出帮助信息。这属于底层工具代码，不构成所声明的核心业务行为，也未展示任何与快手、竞品、视频、博主或评论资源交互的逻辑。因此该代码块与声明用途存在明显不匹配。
The declared purpose describes a feature-rich Kuaishou competitor analysis skill, but the supplied code chunk is only a small utility that reads package.json from the local filesystem and returns the package name. This behavior does not implement or directly support the described user-facing research capabilities in any evident way from the snippet alone. While utility code can be ancillary, the actual behavior shown is materially unrelated to the declared purpose, and it also accesses local files without that being suggested by the description.
The skill declares use of a sensitive environment variable (GUAIKEI_API_TOKEN) but does not define any explicit tool scope or permission boundaries. That creates an overbroad trust model where an agent/runtime may expose secrets or allow execution context beyond what is clearly authorized, increasing the risk of unintended secret access or misuse.
The code reads GUAIKEI_API_TOKEN from the environment and then makes remote API calls to create and query a comment task. For code files, accessing credentials and transmitting user-supplied data are safety-relevant operations that should have some visible disclosure unless clearly documented elsewhere; this file does not include an inline warning, prompt, or explanatory comment about that behavior.
The CLI writes fetched comment data to a local JSON file after completing the request, creating a persistence side effect beyond simple retrieval/display. Because comments can contain personal data, usernames, or sensitive business research artifacts, silent local storage increases data exposure and retention risk on shared machines or downstream systems.
The script stores full comment results locally without any explicit warning or consent in this file, which can surprise users and create unintended data retention. Since this skill is used for competitor and brand research, persisted outputs may contain collected third-party content or analyst work product that should not be silently archived.
The CLI silently writes fetched results to a local file without explicit user notice or consent, which can unexpectedly persist collected third-party content on disk. In a research/data-gathering skill, this increases privacy and data-handling risk, especially on shared machines or when users assume the tool is ephemeral.
The script sends the user-supplied keyword and search parameters to a remote service using an API token, but the user is only told to configure the token, not that their input will leave the local environment. For a research-oriented skill, undeclared external transmission can expose sensitive strategy terms, brand names, or internal investigation topics to a third-party service.
The CLI writes full search output to a local JSON file whose name includes the user's keyword, creating undeclared local persistence of potentially sensitive research activity. In the context of a competitor-research skill, saved queries and results may expose business intelligence, user interests, or collected comment data to other local users, backups, or log collectors.
The script stores search results locally without an explicit warning or consent flow, which can surprise users and retain data longer than expected. Because this skill is used for competitive research, the persisted output may contain sensitive market-intelligence queries and harvested content that should not be silently logged.
This file contains natural-language comments and all user-facing CLI messages exclusively in Chinese, including errors and help output. Under the policy, forcing a specific language without user opt-in or documented locale justification is a natural-language policy violation.
This code contains user-facing status and error strings in Chinese only, such as the validation errors on L07 and L11. The policy for natural-language violations applies to all file types, and there is no indication that the skill offers language selection or that a Chinese-only locale is intentionally required.
This utility performs outbound HTTPS GET/POST requests and attaches a TOKEN credential in request headers, which can transmit user or system data to an external service. In this file there is no confirmation prompt, user-facing logging, or comment/docstring warning about the network transmission or credential use.
The code places the supplied token into the TOKEN HTTP header for outbound requests, which constitutes use of a sensitive credential. This file does not include any visible warning, comment, or user-facing notice that such credentialed requests will be made.
This code emits all validation errors in Chinese string literals, which effectively forces a specific language on users. The file does not provide any opt-in, localization mechanism, or justification that the skill is intended only for a Chinese-speaking or region-specific context.
The manifest description is entirely in Chinese and presents the skill's usage context in that language only. Under the policy, locale or language constraints should either offer user choice or be clearly justified as region-specific; here the skill is Kuaishou-specific, but it does not explicitly state that interaction/output language is restricted by design or offer alternatives.
The package description is written entirely in Chinese and does not indicate any language choice or opt-in. Under the stated policy, natural-language content that forces a specific language without user choice can be considered a locale/language policy violation.
This markdown file presents all user-facing instructions and parameter descriptions exclusively in Chinese. Under the policy rules, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale restriction is clearly documented and justified.
This code sends a user-provided URL to a remote API and includes an authentication token, which is a safety-relevant network operation. Although retry errors are logged, there is no user-facing disclosure, confirmation, or inline documentation here explaining that user data will be transmitted to an external service.
The manifest frames the skill as a Kuaishou competitor-research utility for fetching videos and comments, but does not mention credential handling or dependence on environment-stored secrets. Reading process.env.GUAIKEI_API_TOKEN introduces secret access capability that is not evident from the declared purpose alone.
The manifest frames the skill as a Kuaishou research/fetching tool, but does not mention credential handling or environment-based secret access. Reading process.env.GUAIKEI_API_TOKEN is not inherently malicious, but it is an additional capability beyond the user-facing purpose description.
The manifest describes functionality for obtaining competitor videos, blogger works, and comments for research use, which implies read-oriented retrieval. This CLI additionally persists the full result payload to disk via log.taskWrite, introducing a storage side effect not stated in the manifest description.
The manifest describes competitor-research retrieval functions for Kuaishou content, but does not mention credential handling or environment-based secret access. While authentication may be operationally necessary, this capability is not justified by the stated purpose alone and should be explicitly declared if part of the skill's trust model.
Several error strings are hard-coded in Chinese, including credential and network-related messages, without offering a language choice or documenting a justified locale restriction. This can violate language/locale policy when the skill is not explicitly limited to Chinese-speaking users.
The thrown error string at L21 is hardcoded in Chinese ("重试...次后失败"). This imposes a specific language in user-visible text without any indication of user choice or a documented region-specific requirement, which matches the locale-policy violation criteria.
Detected: suspicious.exposed_secret_literal
