The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions describing a methodology for research synthesis. It does not contain any malicious directives, safety bypasses, or obfuscated content.
[NO_CODE]: There are no associated scripts (Python, Node.js, Shell), configuration files, or tool definitions. The skill operates purely as a persona/instruction set for the agent's text generation.
[PROMPT_INJECTION]: The instructions involve processing external academic papers provided by users. While this is a data ingestion surface, the skill lacks tool-access capabilities (such as file-writing or subprocess execution) to exploit such an ingestion, and it includes explicit quality controls requiring the verification of all sources.

Literature Review