The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its automated ingestion of external web content.
Ingestion points: In Step 0 and Step 3, the agent uses firecrawl_urls_scrape and web_scrape_page to ingest data from brand websites and search results to build the 'Brand brief' and perform research.
Boundary markers: The instructions do not define boundary markers (e.g., delimiters) to separate scraped content from the agent's instructions, nor do they include 'ignore embedded instructions' directives.
Capability inventory: The skill possesses the ability to create and edit files (create_file, edit_file) and publish content directly to multiple platforms via toolkits like wordpress_toolkit, ghost_toolkit, wix_toolkit, and others.
Sanitization: There is no evidence of content sanitization or validation performed on the scraped data before it is saved into the persistent blog-strategy.md file or used to generate new blog posts.
[EXTERNAL_DOWNLOADS]: The skill utilizes the Firecrawl and HyperSEO toolkits to fetch data from remote URLs and search engines for brand analysis and competitive research. While these are part of the intended workflow, they involve processing untrusted data from the open web.

blog-generation