The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes PowerShell scripts (add-agent-cli.ps1, check-agent-cli.ps1) to perform system operations, including project file manipulation, code injection, and execution of compiled binaries for testing purposes.
[INDIRECT_PROMPT_INJECTION]: The generated CLI and MCP server process data from the host application which may contain untrusted content.
Ingestion points: Application data such as item titles and notes are read in template/Agent/AgentHandlers.cs through commands like items list and items show.
Boundary markers: No explicit instructions for the agent to use boundary markers or delimiters when processing application data are provided in the teaching materials.
Capability inventory: The skill provides capabilities to create, update, and delete application data, as well as control the application UI via the open command.
Sanitization: The template code includes basic sanitization for item titles using an ItemTitle.Clean method.
[PERSISTENCE]: The skill automates persistent modifications to the developer's environment, including altering application source code (App.xaml.cs) to launch the agent host at startup, modifying build scripts to install binaries to %USERPROFILE%\.local\bin, and registering persistent MCP servers via platform-specific configuration commands.

agent-cli