The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified in the skill files. All operations are confined to generating text prompts based on local templates and user input.
[PROMPT_INJECTION]: The skill contains no instructions to bypass safety guidelines or override agent constraints. Rule 7 explicitly handles IP (Intellectual Property) names by providing user warnings, which demonstrates a safety-conscious design.
[DATA_EXFILTRATION]: No network-enabled tools or sensitive file path accesses were detected. The skill does not attempt to read credentials, SSH keys, or environment variables.
[COMMAND_EXECUTION]: There is no usage of shell execution or dynamic context injection (the !command syntax). All instructions are purely declarative or logic-based within the prompt generation framework.
[REMOTE_CODE_EXECUTION]: The skill does not download external scripts or packages. It relies on a local directory of Markdown templates which are accessed using standard file-read capabilities.
[CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, or secrets were found in the skill metadata or instructional content.

shortfilm-prompt