The Agent Skills Directory

[SAFE]: The skill implements a robust framework for creating ad content that respects character limits and formatting rules for platforms like Google Ads, Meta, and LinkedIn.
[EXTERNAL_DOWNLOADS]: Reference documentation (references/generative-tools.md) provides information on using established AI generation services including ElevenLabs, Google Gemini, and OpenAI. These resources are standard in the industry and the examples provided use safe practices for API key management.
[COMMAND_EXECUTION]: The skill includes instructions for using command-line tools such as the Google Ads CLI and Remotion. These tools are used for their intended purposes of data retrieval and video rendering within a marketing workflow.
[DATA_EXFILTRATION]: The skill reads local context files (product-marketing-context.md) and processes performance metrics provided by the user. This data access is essential for the skill's functionality and does not involve any suspicious or unauthorized network operations.
[PROMPT_INJECTION]: While the skill ingests external performance data (CSV or text) to iterate on creative, the risk of indirect prompt injection is minimal given the specialized analytical context and the absence of high-privilege tool access tied to the data processing loop.

ad-creative