The Agent Skills Directory

[SAFE]: The skill provides legitimate documentation and code samples for a well-known open-source UI library. Its purpose matches its behavior, focusing entirely on UI component implementation and styling.
[EXTERNAL_DOWNLOADS]: Installation instructions use standard package managers (npm) to fetch the official tdesign-miniprogram library. External URL references in the documentation point to trusted domains belonging to Tencent (gtimg.com, tencent.com) for assets or WeChat developer resources for demonstrations.
[INDIRECT_PROMPT_INJECTION]: The skill provides components for building AI chat interfaces (e.g., t-chat-message, t-chat-content).
Ingestion points: External data enters the agent context through the content property of chat components (e.g., in references/miniprogram-chat/components/chat-content.md).
Boundary markers: Not explicitly defined in the UI component code itself.
Capability inventory: No dangerous system capabilities (exec/eval, file-write, or non-whitelisted network operations) are included in the skill's example scripts.
Sanitization: The components use standard Markdown parsing via the marked library to render content. Given the lack of dangerous capabilities, this represents a low-risk surface area.

tdesign-miniprogram