The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes the gemini command to process user prompts and manage CLI extensions, which is the primary intended behavior of the skill.
[EXTERNAL_DOWNLOADS]: The skill metadata specifies the installation of gemini-cli using Homebrew. This is an established package management service, and the tool is associated with a well-known service provider.
[PROMPT_INJECTION]: The skill accepts natural language prompts as command-line arguments, which creates a surface for potential indirect prompt injection if the agent interpolates untrusted data into the prompts.
Ingestion points: Positional prompt arguments in the gemini "..." commands.
Boundary markers: None identified in the provided usage examples.
Capability inventory: Execution of the gemini binary which interacts with remote AI services and can manage local extensions.
Sanitization: No explicit input sanitization or validation logic is defined within the skill instructions.

gemini