The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute local shell scripts (scripts/generate-film.sh and core/platform/check-result.sh) to perform video generation and status checking operations.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface where untrusted user input is expanded into technical directives and then interpolated into a bash command.
Ingestion points: The subject and intent parameters supplied by the user to the agent as described in the Creative Brief (SKILL.md).
Boundary markers: Absent. The prompt optimization protocol does not specify the use of delimiters or 'ignore' instructions to prevent the model from obeying instructions embedded within the user-provided subject or intent.
Capability inventory: The skill uses bash to execute local scripts that interact with the file system and likely perform network operations to video generation services.
Sanitization: Absent. There are no instructions or mechanisms described to validate, escape, or sanitize the user-provided strings before they are passed as arguments to the bash script.

muapi-cinema-director