The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (dotnet new search, dotnet new list, dotnet new help) using user-provided strings as arguments. This creates a potential command injection vulnerability where an attacker could provide malicious shell metacharacters in the search keywords to execute arbitrary commands on the host system.
Ingestion points: User intent and keywords provided as input in SKILL.md.
Boundary markers: Absent; user input is directly passed to the command line.
Capability inventory: Subprocess execution via dotnet CLI in SKILL.md.
Sanitization: Absent; no instructions are provided to sanitize or escape user-supplied keywords before execution.
[EXTERNAL_DOWNLOADS]: The skill uses the dotnet new search command, which fetches template metadata and package information from the NuGet registry (nuget.org). NuGet is a well-known service for the .NET ecosystem, and this behavior is consistent with the skill's stated purpose of project template discovery.

template-discovery