The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute complex ffmpeg commands for video processing, including text overlays and transitions. It also uses the macOS open command to launch Google Chrome with specific profile directories and URLs for social media platforms.
[EXTERNAL_DOWNLOADS]: The skill connects to Google Gemini API endpoints (generativelanguage.googleapis.com) to generate content, images, and video. It also downloads video assets from URIs returned by these services using urllib.request.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection.
Ingestion points: Reads a brand context file (product-marketing-context.md) and accepts arbitrary user-provided topics and angles.
Boundary markers: Absent. The skill does not define clear delimiters for external data or instructions to ignore embedded commands.
Capability inventory: The skill can execute shell commands via ffmpeg, run Python code for media manipulation and API requests, write files to the local system (generated-assets/, post-history/), and control the web browser via the open command.
Sanitization: No validation or sanitization of ingested content is specified before the data is used in prompts or command templates.

social-posts