The Agent Skills Directory

[SAFE]: The skill is a well-defined set of instructions for the purpose of code review. It contains technical guidelines and criteria for evaluating animation quality and performance.
[PROMPT_INJECTION]: The skill uses instructional language to define a persona and a specific "Operating Posture." These instructions are focused on task performance and do not attempt to override AI safety filters or system-level constraints.
[DATA_EXFILTRATION]: The skill does not access sensitive files, environment variables, or credentials. It contains no commands or scripts for sending data to external network locations.
[REMOTE_CODE_EXECUTION]: There are no patterns involving remote script downloads, package installations, or execution of untrusted code from external sources.
[COMMAND_EXECUTION]: The skill does not use shell commands, subprocess calls, or privilege escalation techniques.
[INDIRECT_PROMPT_INJECTION]: The skill processes code diffs, which is an external data source. However, it lacks capabilities such as file system writes or network access that would be necessary for an indirect prompt injection to perform malicious actions. The output is limited to text-based reviews.

review-animations