The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and process content from external sources (lokalise.com) including blogs, product pages, and webinars. This content is processed by the agent, which could allow instructions embedded in the external web pages to influence the agent's behavior.
Ingestion points: Content is ingested from lokalise.com through tools defined in SKILL.md, such as lokalise_com__read_page and lokalise_com__get_blog_next_d_locale_blog_page_txt.
Boundary markers: There are no explicit boundary markers or instructions to ignore commands embedded in the retrieved text described in the skill instructions.
Capability inventory: The skill provides capabilities to read data and interact with the Unbrowse API to execute tasks on the target site (SKILL.md).
Sanitization: The skill description does not specify any sanitization or filtering logic applied to the external data before it is presented to the agent.

lokalise