Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed Kuaishou public-data research skill that uses a third-party API token, outbound requests, and automatic local result logs.
Install only if you are comfortable sending Kuaishou keywords, creator or video identifiers, request parameters, and the GUAIKEI_API_TOKEN to guaikei.com. Treat the generated logs as potentially sensitive research data because they may contain public comments, author details, queried URLs, and keywords; delete or protect them when no longer needed.
The skill indicates local filesystem writes for log persistence, but that capability is not clearly surfaced in the top-level declaration. Undeclared data persistence is security-relevant because collected comments, author info, and query targets may be written to disk and retained longer than the operator expects, increasing privacy and data-handling risk.
The skill indicates local filesystem writes for log persistence, but that capability is not clearly surfaced in the top-level declaration. Undeclared data persistence is security-relevant because collected comments, author info, and query targets may be written to disk and retained longer than the operator expects, increasing privacy and data-handling risk.
The skill indicates local filesystem writes for log persistence, but that capability is not clearly surfaced in the top-level declaration. Undeclared data persistence is security-relevant because collected comments, author info, and query targets may be written to disk and retained longer than the operator expects, increasing privacy and data-handling risk.
The skill indicates local filesystem writes for log persistence, but that capability is not clearly surfaced in the top-level declaration. Undeclared data persistence is security-relevant because collected comments, author info, and query targets may be written to disk and retained longer than the operator expects, increasing privacy and data-handling risk.
The skill indicates local filesystem writes for log persistence, but that capability is not clearly surfaced in the top-level declaration. Undeclared data persistence is security-relevant because collected comments, author info, and query targets may be written to disk and retained longer than the operator expects, increasing privacy and data-handling risk.
The skill indicates local filesystem writes for log persistence, but that capability is not clearly surfaced in the top-level declaration. Undeclared data persistence is security-relevant because collected comments, author info, and query targets may be written to disk and retained longer than the operator expects, increasing privacy and data-handling risk.
The skill declares access to an environment secret (GUAIKEI_API_TOKEN) but does not define any explicit tool scope or permission boundary. That weakens least-privilege guarantees and makes it harder for the host agent/runtime to constrain what the skill may access, especially if the implementation later grows to include broader file, network, or subprocess behavior.
The skill is ներկայացված as a Kuaishou research and competitive-analysis tool, but its package scripts expose active posting and commenting capabilities. This expands the operational scope from passive analysis to account actions that could be used for spam, unauthorized engagement, or automation abuse if invoked by an agent or user expecting read-only behavior.
The skill metadata frames the capability as keyword search/detail fetching for competitor research, but the documentation additionally exposes bulk comment collection for specific videos. This expands the skill into user-generated interaction harvesting, which can enable profiling, sentiment scraping, or collection of personal data beyond the declared purpose, reducing transparency and increasing privacy/compliance risk.
The manifest describes a 快手 search/detail fetcher for competitor research, which suggests retrieving search results or item details. This CLI instead creates and retrieves comment tasks and saves the collected comments to disk, which is a materially different data collection behavior than search/detail fetching.
The script writes fetched comment data to a local JSON file via log.taskWrite without any explicit opt-in, warning, retention control, or masking. Because comments can contain personal data, sensitive text, or regulated content, silent persistence increases the risk of unintended data retention, secondary disclosure, and misuse on shared systems or CI runners.
The calls to post.createPostTask and post.getPostTask send the user-provided URL, sort, and limit together with an API token to external request-handling code. Although network access may be part of the tool's purpose, this file's visible help and inline text do not explicitly warn users that their input will be sent to a remote service.
The script persists fetched results to a local JSON file automatically after successful execution, but it does not clearly warn the user that potentially sensitive research data and queried profile identifiers will be stored on disk. In shared workstations, CI runners, or multi-user environments, this can lead to unintended retention and later disclosure of collected data.
This JavaScript file contains natural-language comments and user-visible CLI messages entirely in Chinese, including errors and help output. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale restriction is explicitly justified, which is not present here.
Multiple user-visible error messages are hard-coded in Chinese, including credential and request-failure messages. This imposes a specific language on users without offering a locale choice or documenting a justified region-specific constraint.
This code file defines generic GET and POST helpers that send caller-supplied params, JSON data, and a token to a remote host via HTTPS. There is no confirmation prompt, logging, comment, or docstring disclosing that user or system data may be transmitted, so the network operation lacks visible warning in this file.
The helper inserts the provided token into the TOKEN header for outbound requests, which is a credential-handling operation. In this file there is no comment, docstring, or user-facing notice explaining that credentials are being transmitted to an external service.
This code emits all validation errors exclusively in Chinese via user-facing string literals. That imposes a specific language on users without offering any locale selection or documenting a justified region-specific constraint, which matches the language/locale policy violation criteria.
The package description is written entirely in Chinese and presents the skill as a general-purpose professional analytics tool, but there is no indication that the Chinese locale is optional or that the skill is intentionally restricted to Chinese-speaking users. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation.
The README states that all task results are automatically saved to the logs/ directory, but it does not clearly warn users that collected platform data, URLs, keywords, and potentially comment content will be persisted to disk. In a data-collection tool used for competitor research and comment analysis, silent local retention increases the risk of unintended exposure through shared machines, backups, source-control commits, or insecure log handling.
This markdown file is entirely written in Chinese, including the title and changelog entries, with no indication that users may choose another language or that the skill is intentionally limited to a Chinese-speaking audience. The policy specifically calls for flagging language or locale constraints when they are imposed without user opt-in or clear justification.
This markdown file presents all usage instructions and parameter descriptions only in Chinese, which can functionally force a specific language on users. The policy allows locale constraints only when users are given a choice or the restriction is clearly documented and justified, neither of which appears here.
The error message string is hard-coded in Chinese, which indicates the skill presents output in a specific language without offering a language or locale choice. This can violate language/locale policy when no opt-in or justification is provided in the file.
This thrown error message is also hard-coded in Chinese, reinforcing that the skill's natural-language output is fixed to one language. Without any evidence of user choice or documented locale scope, this is a policy concern.
The retry log message is another user-visible string fixed to Chinese, suggesting the skill enforces a specific language by default. The file shows no mechanism for locale configuration or user opt-in.
Detected: suspicious.exposed_secret_literal
