The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill involves a data ingestion surface where the agent is instructed to crawl external websites and process untrusted data.
Ingestion points: The "Check" section in SKILL.md and "Finding Duplicates" in references/rule.md instruct the agent to crawl all pages and extract <title> tag content.
Boundary markers: Absent. There are no instructions to delimit the extracted title content or treat it as untrusted data.
Capability inventory: The skill implies the use of web crawling tools to fetch data, which is then processed to generate audit reports.
Sanitization: Absent. The skill does not specify any sanitization or filtering for the content extracted from external tags.

title-unique