The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the google-generativeai Python package. This is the official SDK provided by Google for interacting with Gemini models and is a standard requirement for this functionality.
[COMMAND_EXECUTION]: The skill defines a workflow utilizing MCP tools generate_creative_image and expand_to_stories. These tools are used for image generation and outpainting, which is the primary purpose of the skill.
[DATA_EXFILTRATION]: The skill accesses local files located in .claude/ads-agent/config/ to retrieve account information and advertisement briefs. This data is used to populate parameters for the image generation tool, which sends data to Google's API as intended by the user request.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes external data from Markdown brief files.
Ingestion points: .claude/ads-agent/config/briefs/{account}.md and user-provided text.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the interpolation logic.
Capability inventory: Includes image generation (generate_creative_image) and file reading (Read tool).
Sanitization: There is no evidence of sanitization or validation of the content read from the brief files before it is passed to the generation tool.

creative-image-generator