Security audit
Security checks for vulnerabilities and agentic risk
The skill is a disclosed content-collection workflow that writes reports to Obsidian and posts summaries to a configured channel, with privacy-sensitive behavior users should configure carefully.
Before installing, confirm the Obsidian Vault path, whether the Vault syncs to cloud storage, and exactly which group webhook or callback URL is configured. Use a dedicated X/browser session if possible, and avoid adding sensitive private feeds or credentials to the configuration.
The paid-feature section advertises '复杂工作流可视化编排', '条件分支与异常重试', '定时触发与事件驱动', and '执行日志与审计追踪' as capabilities. These are generic automation-platform features rather than obvious requirements of a skill whose stated purpose is collecting and ranking X content for creators, creating an unjustified expansion of capability scope.
The skill explicitly writes generated reports into the user's Obsidian Vault, but the documentation does not prominently warn that local files and directories will be created or modified. In an agent setting, silent file writes can surprise users, overwrite content, or persist sensitive scraped material into synced notes without clear consent.
The skill sends collected content to a group channel/webhook but does not prominently warn users that data will be transmitted to an external destination. Because scraped content, mentions, metadata, and possibly user-derived context may be pushed outside the local environment, this creates a meaningful privacy and data-exfiltration risk if users do not realize external transmission occurs.
The display name, summary, and the rest of the skill instructions are presented entirely in Chinese, indicating a fixed language/locale experience. SQP-3 applies to all file types when a skill forces a specific language without user opt-in, and the document does not mention any language selection or alternative locale support.
The manifest describes a content-collection, ranking, channel-push, and Obsidian-write workflow, but the input schema additionally advertises an arbitrary callback_url parameter. Exposing generic callback behavior is not otherwise justified or described as part of this skill's purpose, and implies outbound notification capability beyond the specific 'push to group channel' scope in the manifest.
No suspicious patterns detected.
