The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (scripts/blog_preflight.py, scripts/generate_hero.py, scripts/blog_render.py) to manage article lifecycle tasks like rendering and asset pre-checks. It also utilizes curl in a shell pipeline to verify the HTTP status of external image URLs from providers like Pixabay.
[EXTERNAL_DOWNLOADS]: The skill performs automated research by querying web sources for statistics, images (Pixabay, Unsplash, Pexels), and YouTube video metadata to enrich the blog content.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by integrating unvalidated research data and statistics from external web sources into its drafting workflow.
Ingestion points: Phase 2 research data, including statistics, source URLs, and YouTube metadata.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore potential instructions embedded in the external research content.
Capability inventory: The workflow involves shell command execution (curl), local script execution (Python), and network-enabled research (WebSearch).
Sanitization: There is no explicit description of sanitization or validation logic applied to the research data before it is interpolated into the generated blog post sections.

blog-write