Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed Kuaishou public-data research tool that uses a Guaikei API token, sends user-provided search targets to its API, and saves results locally.
Before installing, confirm you are comfortable sending Kuaishou search terms, profile or video URLs, request parameters, and your Guaikei API token to www.guaikei.com. Also review or delete the generated logs if the research targets or fetched comments are sensitive to your team.
Local filesystem writes and log creation can create data-retention and unintended disclosure risks, especially because the skill handles scraped comment content and associated metadata. In an agent environment, silent persistence may expose collected data to other local processes, later prompts, backups, or users who were not expecting storage.
Local filesystem writes and log creation can create data-retention and unintended disclosure risks, especially because the skill handles scraped comment content and associated metadata. In an agent environment, silent persistence may expose collected data to other local processes, later prompts, backups, or users who were not expecting storage.
Local filesystem writes and log creation can create data-retention and unintended disclosure risks, especially because the skill handles scraped comment content and associated metadata. In an agent environment, silent persistence may expose collected data to other local processes, later prompts, backups, or users who were not expecting storage.
Local filesystem writes and log creation can create data-retention and unintended disclosure risks, especially because the skill handles scraped comment content and associated metadata. In an agent environment, silent persistence may expose collected data to other local processes, later prompts, backups, or users who were not expecting storage.
Local filesystem writes and log creation can create data-retention and unintended disclosure risks, especially because the skill handles scraped comment content and associated metadata. In an agent environment, silent persistence may expose collected data to other local processes, later prompts, backups, or users who were not expecting storage.
The skill declares access to a sensitive environment variable (GUAIKEI_API_TOKEN) but does not define any explicit tool scope, permissions, or allowed-tools boundary. This weakens least-privilege guarantees and makes it harder for a host agent to constrain what the skill may access or invoke, increasing the blast radius if the skill is misrouted, extended, or composed with other capabilities.
The manifest description is entirely in Chinese and presents the skill as generally applicable to content research and operations users, but it does not state that the skill is intended only for Chinese-speaking users or offer an opt-in language choice. Under the policy, forcing a specific language or locale without user choice is a natural-language policy concern.
This manifest file describes the skill in broad marketing terms as a professional Kuaishou data analysis tool, but it does not define specific invocation phrases, contexts, or exclusions. In a manifest context, the absence of precise trigger scope can contribute to ambiguous matching and unintended invocation for general growth or marketing-related requests.
The README states that all task results are automatically saved to the logs/ directory, but it does not clearly warn users that fetched content, query terms, target profile/video URLs, and derived outputs will be written to local storage. For a data-collection skill used in competitor monitoring and comment analysis, this can create unintended retention of potentially sensitive business research data and increase exposure if the host is shared, backed up, or compromised.
This markdown file presents all usage instructions and parameter descriptions only in Chinese, which can amount to a language-policy issue if the organization expects skills not to force a locale without user opt-in. There is no note that the skill is intentionally limited to Chinese-speaking users or a specific region.
The script sends the user-supplied video URL or ID and uses an API token to fetch comments from a remote service, but it provides no explicit notice that data is being transmitted externally. In a research-oriented skill, lack of transparency about outbound data flows can cause users to disclose targets or research activity to a third party without informed consent.
The CLI persists the full fetched comment payload to a local JSON file without making that side effect explicit to the user. Comment data can contain personal or sensitive content, and silent persistence increases the risk of unintended retention, later disclosure, or collection by other local processes.
The manifest name includes "comment-creator-fetcher", suggesting comment retrieval capability, but this CLI creates and retrieves a post task via createPostTask/getPostTask and labels outputs as "获取作品" and "post". That indicates the implemented behavior is creator post fetching, not comment fetching, which is a semantic mismatch in the skill's described intent versus actual operation.
The calls to post.createPostTask and post.getPostTask send the provided URL, sort, limit, and token-derived authentication to an external service, which is a network operation covered by this rule. The script reports task progress, but it does not clearly warn the user that their supplied data will be sent over the network to a remote API.
The code saves the full output payload to disk without an explicit warning or confirmation at the point of persistence. Because the saved JSON includes request metadata and fetched results, it may unintentionally retain potentially sensitive operational or research data beyond the user's immediate session.
This code includes user-facing error strings, comments, and help output entirely in Chinese, such as parameter errors and the generated help text. Because the file does not offer any language/locale choice or document that the tool is intentionally limited to Chinese users, it violates the natural-language policy on forcing a specific language without user opt-in.
This code emits user-facing error messages in Chinese (日志文件名必须是非空字符串, 日志内容必须是非空字符串) and later success/error messages are also Chinese-only. That imposes a specific language on users without any visible opt-in or justification, which matches the language/locale policy violation criteria.
The success and failure output strings (已保存到, 日志写入失败) are presented only in Chinese, with no indication that users can choose another language. Forcing a single locale in user-visible output without opt-in is a natural-language policy concern.
This code performs outbound HTTPS GET and POST requests and includes a TOKEN header, query parameters, and JSON payloads, which may transmit user or system data to a remote service. The file contains no confirmation prompt, logging, comment, or docstring disclosing that network transmission occurs.
The functions require a token and place it into the outbound TOKEN header for remote requests. While token use may be functionally necessary, this file provides no user-visible warning, comment, or documentation indicating that credentials are consumed and transmitted to an external service.
This code performs an outbound POST request that transmits a user-supplied URL along with an authentication token, but the file provides no confirmation prompt, warning, or explanatory comment about that data transmission. Because network calls that transmit user or system data are safety-relevant, the lack of disclosure is noteworthy.
The retry message string is hard-coded in Chinese, which can indicate a language/locale constraint without offering the user any choice. The policy requires avoiding forced language behavior unless it is explicitly justified or user-selected.
This function makes an outbound GET request that includes the provided URL and token, but the file does not include any user-facing notice or inline explanation that this information is sent to an external service. Under the missing-warning rule for code files, transmitting user or system data over the network without disclosure should be flagged.
This thrown error uses a Chinese-only message and the file does not indicate that the skill is intentionally limited to Chinese-language users or that the language is user-selectable. That creates a potential language policy violation for user-visible natural-language output.
The retry status output is fixed in Chinese and there is no evidence in this file that users can choose their preferred language or that a Chinese locale is required. This may violate the policy against forcing a specific language without opt-in.
Detected: suspicious.exposed_secret_literal
