The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-supplied data, such as prompts, scripts, and external media URLs, and interpolates them into okou generate command arguments.
Ingestion points: In SKILL.md, the generation workflow and common patterns sections describe using flags like --prompt, --script, --image-url, and --video-url with user-provided content.
Boundary markers: The skill lacks explicit instructions or delimiters to isolate user-supplied content from the command structure, increasing the risk that a malicious prompt could attempt to manipulate CLI behavior.
Capability inventory: The skill utilizes the okou generate command for artifact creation and okou host for deploying static HTML artifacts (websites, reports, dashboards) authored by the agent.
Sanitization: No sanitization, escaping, or validation logic is defined for the external data before it is passed to the shell commands.
[COMMAND_EXECUTION]: The skill's primary function is to execute the okou CLI tool. As this is the vendor's own provided tool and is central to the skill's purpose, this behavior is expected and operates within the designated scope.