tlysanhuo/aminer-daily-paper

Registry 信息

项目简介

一个基于AMiner、arXiv和LLMs驱动的个性化论文推荐系统，专为OpenClaw和飞书设计，提供定制化的学术论文推荐。

要点

• 集成AMiner、arXiv和LLMs进行论文推荐
• 专为OpenClaw和飞书平台设计
• 基于用户兴趣的个性化推荐

使用场景

• 研究人员的学术论文发现
• 飞书中的个性化阅读列表
• 跟踪最新arXiv论文动态

README 摘要

# aminer-rec Public repository for the `aminer-rec5` OpenClaw / Feishu paper recommendation skill, with two smooth entry points: - scholar bootstrap: start from `aminer_user_id`, `scholar + org`, representative papers, or a local `papers_file` - topic bootstrap: start from `topics` or free-form natural language such as "I work on multimodal agents and tool use" The repository turns both paths into one unified `ResearchProfile`, then runs retrieval, AMiner enrichment, summarization, card rendering, and delivery. The OpenClaw command name remains `/aminer-rec5`. ## Why This Version Exists This is the public-shareable repo cut of `aminer-rec5`: - secrets removed - local output artifacts removed - hard-coded machine paths replaced with portable defaults - README and setup flow rewritten for external users If you want to post this on social platforms and drive traffic to a personal GitHub repo, this version is the one to publish. ## Highlights - Natural-language-first paper recommendation - Scholar-aware cold start from AMiner person signals - Unified topic and scholar profile building - arXiv retrieval plus AMiner enrich - Structured summaries and recommendation reasons - Feishu / OpenClaw-friendly output format - Graceful degradation when optional internal components are unavailable ## Quick Start ### 1. Install dependencies ```bash python3 -m venv .venv source .venv/bin/activate pip install -r requirements.txt ``` ### 2. Create your config ```bash cp config.example.yaml config.yaml ``` You can either fill `config.yaml` explicitly, or leave some fields empty and let the runtime discover them from local OpenClaw / environment settings. - `aminer.token`: your own AMiner token - `llm.api_key`: your OpenAI-compatible model key - `llm.base_url` and `llm.model`:

话题

暂无话题