The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The documentation directs users to install the Syncfusion.Maui.Chat package from the NuGet registry. This is a standard procedure for integrating third-party UI components in .NET development and does not represent a security risk when using established vendors.
[COMMAND_EXECUTION]: The SKILL.md quick start guide includes the dotnet add package command. This is the standard command-line interface method for adding dependencies to a .NET project.
[INDIRECT_PROMPT_INJECTION]: The skill implements a user interface for displaying messages which, by design, ingests and processes content from external and potentially untrusted sources (e.g., other chat participants or automated bots). This establishes a vulnerability surface for indirect prompt injection if the agent interprets instructions embedded within the chat history.
Ingestion points: The Messages collection (documented in SKILL.md) and the ItemsSource property (documented in references/data-binding.md) serve as the entry points for untrusted data into the agent's view.
Boundary markers: There is no specific mention of using delimiters or clear instructions to the agent to treat chat content as data rather than instructions.
Capability inventory: The skill provides a range of interactive event handlers for message submission, card interaction, and suggestion selection (references/events.md), which could be triggered by interactions influenced by injected content.
Sanitization: The provided documentation does not detail any mechanisms for escaping or sanitizing the content of messages before they are rendered or processed by the agent.

syncfusion-maui-chat