The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external search results which could contain malicious instructions aimed at influencing the agent's behavior during newsletter generation or social media posting.
Ingestion points: Web search results are retrieved from tavily/search-assistant and exa/search as described in SKILL.md.
Boundary markers: The skill lacks explicit delimiters or instructions to the agent to ignore instructions embedded within the retrieved search content before processing it into a newsletter template.
Capability inventory: The skill utilizes the belt CLI via Bash to perform downstream operations such as creating social media posts (x/post-create) and generating images from HTML.
Sanitization: No sanitization, filtering, or validation of the content retrieved from external search providers is implemented prior to interpolation into the newsletter structure.

newsletter-curation