The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and act upon data from external, potentially attacker-controlled websites.
Ingestion points: In 'Step 0', the skill uses firecrawl_urls_scrape and web_pages_scrape to read a brand's homepage, about page, and existing blog posts to populate a blog-strategy.md file.
Boundary markers: The instructions lack explicit delimiters or 'ignore embedded instructions' warnings for the data being scraped, increasing the risk that instructions hidden on a website could influence the agent's behavior.
Capability inventory: The skill has significant capabilities, including file system modification (create_file, edit_file) and the ability to publish content to external CMS platforms (WordPress, Ghost, Wix, Webflow).
Sanitization: There is no mention of sanitizing, escaping, or validating the external content before it is processed or appended to the stateful log.

blog-generation