The Agent Skills Directory

[SAFE]: The skill is purely informational and educational. It focuses on mapping the research landscape of Code LLMs.
[EXTERNAL_DOWNLOADS]: The skill includes a Python script that utilizes the arxiv library to fetch paper metadata from the official ArXiv repository. This is a well-known academic service and its use here is consistent with the skill's primary purpose of academic research.
[PROMPT_INJECTION]: No instructional overrides, jailbreak attempts, or commands aimed at bypassing safety filters were detected in the prompt text or metadata.
[DATA_EXFILTRATION]: No patterns associated with credential harvesting, sensitive file access, or unauthorized data transmission were found.
[COMMAND_EXECUTION]: The skill does not contain shell commands, subprocess calls, or mechanisms for arbitrary code execution.
[INDIRECT_PROMPT_INJECTION]: While the skill processes external data (arXiv search results), it lacks any high-privilege capabilities (such as file system write access or system command execution) that could be exploited by malicious content within paper titles or abstracts.

code-llm-papers-guide