The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute shell commands based on user input. The profile name extracted from $ARGUMENTS is placed directly into the command gsd-sdk query config-set-model-profile <profile-name> --raw. Without sanitization, this pattern is vulnerable to command injection where an attacker could append additional commands (e.g., using ; or &&) to be executed by the agent.
[INDIRECT_PROMPT_INJECTION]: The skill exposes an attack surface by processing untrusted input that influences tool behavior.
Ingestion points: User-provided $ARGUMENTS in the SKILL.md file.
Boundary markers: None identified in the provided instructions.
Capability inventory: The skill has access to Bash, Read, and Write tools, allowing for system modification and file access.
Sanitization: There is no evidence of input validation or escaping of shell metacharacters before the user input is used in the gsd-sdk command execution.

gsd-config