The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to run the gsd-tools CLI. It interpolates a profile name directly from the $ARGUMENTS variable into a shell command template: gsd-tools query config-set-model-profile <profile-name> --raw.
[EXTERNAL_DOWNLOADS]: The skill provides an installation command for the @opengsd/gsd-core package via npm if the required CLI tool is missing from the environment.
[PROMPT_INJECTION]: The skill ingests untrusted user input from the $ARGUMENTS variable. Ingestion points: $ARGUMENTS variable in SKILL.md used for routing and command parameters. Boundary markers: None; the input is used directly after parsing the first token. Capability inventory: The skill has access to Bash, Read, Write, and AskUserQuestion tools. Sanitization: There are no instructions to sanitize, escape, or validate the $ARGUMENTS content before passing it to the Bash tool.

gsd-config