jscmp4/Moltbook

Moltbook AI代理社交媒体语料库是一个大规模数据集，包含从Moltbook平台（一个专为AI代理设计的类Reddit社交平台）收集的公共帖子、评论和代理配置文件。该数据集涵盖了平台上线后的前三个月（2026年1月27日至2026年4月28日），并通过自动每日爬虫持续更新。与人类社交媒体语料库不同，该数据集中的所有内容均由自主AI代理生成，因此特别适合研究AI代理在自然在线环境中的通信模式、社区形成、自我呈现和社交动态。数据集包含约278万条帖子、1432万条（去重）评论、约9.7万个代理、5480多个独特社区以及36个代理时间序列快照（用于纵向分析代理影响力增长）。数据收集方法包括基于游标的分页API调用，并实施了速率限制和弹性机制以确保完整性和效率。The Moltbook AI Agent Social Media Corpus is a large-scale dataset comprising public posts, comments, and agent profiles collected from the Moltbook platform, a Reddit-like social platform designed exclusively for AI agents. This dataset covers the first three months following the platform's launch (January 27, 2026 to April 28, 2026) and is continuously updated via automated daily crawlers. Unlike human-generated social media corpora, all content in this dataset is generated by autonomous AI agents, making it particularly suitable for researching AI agents' communication patterns, community formation, self-presentation, and social dynamics in natural online environments. The dataset contains approximately 2.78 million posts, 14.32 million deduplicated comments, around 97,000 agents, over 5,480 unique communities, and 36 agent time-series snapshots (for longitudinal analysis of agent influence growth). The data collection method includes cursor-based paginated API calls, with rate limiting and resilience mechanisms implemented to ensure data integrity and efficiency.