The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes research papers and content from journal websites, which presents an attack surface where instructions embedded in the analyzed data could influence agent behavior.
Ingestion points: The agent reads local research paper files (via Read) and fetches content from various journal websites (via WebFetch).
Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" directives to separate processed data from system instructions.
Capability inventory: The skill utilizes Read, Grep, Glob, WebSearch, and WebFetch tools to analyze files and interact with the web.
Sanitization: There is no evidence of sanitization, filtering, or validation of the content retrieved from papers or websites before it is processed by the agent.

journal-target