The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection by ingesting and processing data from external sources.
Ingestion points: The skill reads external data from Naver search results and AI Briefing sources via the check_naver_visibility tool, and retrieves existing blog content via the open_post tool.
Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from potentially following instructions that might be embedded within the retrieved search results or existing blog posts.
Capability inventory: The skill has the ability to perform write operations by updating blog posts through the update_post tool and by handing off content to the naver-post skill.
Sanitization: The instructions do not specify any sanitization, filtering, or escaping mechanisms for the external content before it is used to generate or rework blog drafts.

naver-ai-cite