The Agent Skills Directory

[PROMPT_INJECTION]: The instructions are strictly focused on identifying motion terms and providing animation recipes. No bypass attempts, role-play injections, or instructions to ignore safety guidelines were detected.
[DATA_EXFILTRATION]: The skill does not contain any network operations, hardcoded credentials, or access to sensitive file paths. Mentions of 'downloaded effects' in the audio sync file are conceptual descriptions of a workflow rather than active commands.
[OBFUSCATION]: The markdown content is clear and readable. No Base64, zero-width characters, homoglyphs, or steganographic patterns were found.
[REMOTE_CODE_EXECUTION]: There are no scripts or commands that download or execute external code. The skill is entirely composed of informational markdown files.
[COMMAND_EXECUTION]: References to local tools like 'motion-studio' are included as part of the domain-specific workflow descriptions and do not involve executable shell commands or scripts within the skill package.
[INDIRECT_PROMPT_INJECTION]: Although the skill processes user-supplied 'director notes,' it serves as a static lookup table. It lacks the necessary capabilities (such as network access or file system modification) to be leveraged for an injection attack.

motion-vocabulary