The Agent Skills Directory

[Remote Code Execution Pattern]: The skill references an installation method for the Aspire CLI using curl -sSL https://aspire.dev/install.sh | bash. While piping remote scripts to a shell is a security consideration, this pattern points to the official product domain for Microsoft Aspire and is a documented installation path for the tool.
[External Package Installation]: The documentation suggests installing the Aspire CLI via global managers like npm install -g @microsoft/aspire-cli and dotnet tool install -g Aspire.Cli. These packages are hosted on official registries (NPM and NuGet) and use the author's verified namespace.
[Command Execution]: The skill executes several CLI commands (aspire init, aspire new, aspire doctor) to perform project bootstrapping and environment diagnostics. These operations are performed with the --non-interactive flag to ensure predictable agent behavior.
[Indirect Prompt Injection Surface]: The skill's detection logic relies on scanning the local repository for specific file patterns (e.g., .csproj files with Aspire.AppHost.Sdk or aspire.config.json).
Ingestion points: Filesystem metadata and file contents are used to decide between aspire new and aspire init workflows (SKILL.md).
Boundary markers: The logic is constrained to specific filenames and string matches, reducing the risk of processing unintended data.
Capability inventory: The skill has the capability to write to the filesystem and execute shell commands through the Aspire CLI.
Sanitization: The skill uses explicit flags like --non-interactive and --language to limit the execution scope and prevent ambiguous command outcomes.

aspire-init