The Agent Skills Directory

[SAFE]: The skill processes external documents including PDFs and URLs to perform its core summarization function. Although this constitutes an ingestion point for untrusted data, the skill uses explicit labeling (e.g., '【论文原文】', '【相关文献】') to maintain a clear boundary between the source text and the agent's analysis.
[SAFE]: The generation of interactive HTML components and code snippets for reproduction experiments are legitimate features used for pedagogical purposes. These components are self-contained, and the skill includes fallback mechanisms for environments where execution is restricted.
[SAFE]: No obfuscation, data exfiltration patterns, or persistence mechanisms were detected. The technical workflow follows established research study patterns.

summarize-paper