The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown-based instructions for prompt engineering and does not include any executable scripts, binaries, or automated tool configurations.
[DATA_EXPOSURE_AND_EXFILTRATION]: No hardcoded credentials, sensitive file paths, or data exfiltration patterns were detected. The instructions specifically state that API keys should never be included in the generated prompt.
[INDIRECT_PROMPT_INJECTION]: While the skill is designed to process untrusted external data (such as user stories and media references), it mitigates potential injection risks by providing strict templates and using clear boundary markers (e.g., curly braces for dialogue, brackets for subtitles) to wrap external content. The skill lacks exploitable capabilities like shell execution or network access.
[METADATA_POISONING]: The metadata fields (name, description) accurately reflect the skill's behavior and do not contain deceptive or hidden instructions.

sd25-pe