← Back to Skills Marketplace
kingsunzhang2026-oss

学术论文检索小助手

by viflow · GitHub ↗ · v1.0.2 · MIT-0
cross-platform ✓ Security Clean
46
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install songge-academic-search
Description
学术论文检索小助手(松哥版)。支持多源检索(OpenAlex/Semantic Scholar/Crossref/arXiv/PubMed),自动补全论文元数据,输出 BibTeX/RIS/JSON/Markdown 引用格式。当用户搜索学术论文、查找文献、生成参考文献列表时触发。
README (SKILL.md)

学术论文检索小助手

专业学术论文多源检索工具,支持 OpenAlex、Semantic Scholar、Crossref、arXiv、PubMed 五大数据源,自动 enrichment 补全元数据。

作者:松哥(Zhang JinSong)


核心功能

  • 🌐 五库并行检索(OpenAlex / Semantic Scholar / Crossref / arXiv / PubMed)
  • 🔗 Multi 模式自动串联:OpenAlex 搜 → S2 补摘要/引用 → Crossref 补卷期页
  • 📤 全格式导出:Text / JSON / BibTeX / RIS / Markdown
  • 📥 PDF 下载(arXiv、OpenAlex 有 PDF 时)
  • 🔑 S2_API_KEY 完全可选(无 key 正常运行,仅受速率限制)

数据源说明

数据源 是否需要 Key 特点
OpenAlex ❌ 免费 覆盖全学科,2 亿+ 记录,完整元数据
Semantic Scholar ⚠️ 可选 CS 领域最全,有 key 速率更快(无 key 限 1/s)
Crossref ❌ 免费 补充期刊卷/期/页/abstract
arXiv ❌ 免费 预印本,PDF 可下载
PubMed ❌ 免费 生物医学方向

Semantic Scholar API Key 配置说明

完全可选。 不配置也能正常使用,仅受速率限制(约 1 次/秒)。

有 key 的效果: 解除速率限制,每秒可发更多请求。

申请步骤:

  1. 访问 https://www.semanticscholar.org/product/api
  2. 注册账号,申请 Free tier(免费)
  3. 使用时通过 CLI 参数传入:--semantic-api-key 'your-key'

示例:

# 无 key(正常运行)
python scripts/research.py multi "transformer attention" -n 10

# 有 key(速率更快)
python scripts/research.py multi "transformer attention" -n 10 --semantic-api-key 's2k-xxxxx'

注意: 本 skill 不读取也不存储 S2_API_KEY 环境变量,Key 仅通过 CLI 参数传入。


创作流程

Step 1:确认检索需求

  • 检索主题/关键词
  • 目标数据库(multi 推荐)
  • 结果数量

Step 2:执行检索

  • 推荐使用 multi 模式(自动三步串联)
  • 或指定单一数据源(openalex / semantic / arxiv / pubmed)

Step 3:选择输出格式

  • 日常阅读 → text
  • 论文写作 → bibtex(LaTeX / Word 插件导入)
  • 文献管理器 → ris(Zotero / Mendeley 导入)
  • 结构化数据 → json

命令行使用

python scripts/research.py \x3Csource> "\x3C关键词>" [选项]

数据源(source)

source 说明 是否需要 Key
openalex 推荐,全学科覆盖
semantic Semantic Scholar ⚠️ 可选
multi 自动串联三库(推荐) ⚠️ S2 可选
crossref 用 DOI 查元数据
arxiv 预印本 + PDF 下载
pubmed 生物医学文献

常用选项

# 推荐:multi 模式(自动补全)
python scripts/research.py multi "large language model alignment" -n 10 -f bibtex

# OpenAlex 搜索(无需 key)
python scripts/research.py openalex "graph neural networks" -n 20 --year 2023

# 带引用数过滤的 S2 搜索(有 key 更快)
python scripts/research.py semantic "reinforcement learning" -n 10 --min-citations 100

# arXiv 下载 PDF
python scripts/research.py arxiv "transformers" -n 5 --download --output-dir ./papers/

# PubMed 生物医学检索
python scripts/research.py pubmed "CRISPR gene editing" -n 10

# 生成 BibTeX
python scripts/research.py multi "attention mechanism" -n 20 -f bibtex -o refs.bib

完整参数说明

参数 说明 适用数据源
-n, --max-results 最大结果数(默认 10) 全部
-f, --format 输出格式(text/json/bibtex/ris/markdown) 全部
-o, --output 保存到文件 全部
--year 年份过滤 openalex/semantic/arxiv
--start-date 开始日期(YYYY-MM-DD) openalex/pubmed
--end-date 结束日期 openalex/pubmed
--author 作者名 openalex/semantic/arxiv/pubmed
--min-citations 最低引用数 semantic
--enrich-s2 S2 补全详情(摘要/引用数) semantic
--crossref-enrich Crossref 补充元数据(默认开启) openalex/multi
--journal-issn 按 ISSN 筛选期刊 openalex
--concept-id 按学科概念 ID 筛选 openalex
--category arXiv 学科分类(如 cs.LG) arxiv
--publication-type PubMed 文献类型 pubmed
--doi-file 从文件读取 DOI 列表 crossref
--download 下载 arXiv PDF arxiv
--output-dir PDF 下载目录 arxiv
--sort-by 排序(relevance/date/citations) openalex

参考文档

文件 内容 何时读取
references/readme.md 详细使用说明、Workflow 示例 检索前必读
references/readme.md 安装依赖、故障排查 配置环境时

输出格式示例

BibTeX(用于 LaTeX / Zotero)

@article{vaswani2017attention,
  title={Attention Is All You Need},
  author={Vaswani, Ashish and Shazeer, Noam and Parmar, Niki},
  year={2017},
  journal={arXiv preprint},
  doi={10.48550/arXiv.1706.03762},
  url={https://arxiv.org/abs/1706.03762}
}

RIS(用于 Zotero / Mendeley)

TY  - JOUR
TI  - Attention Is All You Need
AU  - Vaswani, Ashish
AU  - Shazeer, Noam
PY  - 2017
JO  - arXiv preprint
DO  - 10.48550/arXiv.1706.03762
ER  -

安装依赖

pip install -r scripts/requirements.txt

注意: S2_API_KEY 环境变量由用户在 ~/.bashrc 中自行配置,不写入 skill 文件。


MIT License · 松哥专版

Usage Guidance
This skill appears safe for normal academic-search use. Before installing, use a virtual environment for the Python dependencies, keep any Semantic Scholar API key private, and remember that your search queries and DOI lookups are sent to external scholarly services.
Capability Analysis
Type: OpenClaw Skill Name: songge-academic-search Version: 1.0.2 The skill bundle is a legitimate academic research tool that integrates multiple scholarly data sources including OpenAlex, Semantic Scholar, Crossref, arXiv, and PubMed. The Python script `scripts/research.py` uses standard libraries to perform API queries and format bibliographic data (BibTeX, RIS, etc.), with no evidence of data exfiltration, unauthorized file access, or malicious command execution. The documentation in `SKILL.md` and `references/readme.md` provides clear instructions for the AI agent and users, including transparent handling of optional API keys and a user-triggered PDF download feature.
Capability Tags
requires-sensitive-credentials
Capability Assessment
Purpose & Capability
The search, enrichment, citation export, and optional PDF download features are coherent with the stated academic-search purpose, with normal external API data flow.
Instruction Scope
The workflow is user-triggered and scoped to research queries, but the Semantic Scholar key instructions are slightly inconsistent between CLI-only use and optional shell-profile configuration.
Install Mechanism
There is no automatic install spec, but the docs instruct a user-directed pip install with non-pinned dependency ranges.
Credentials
Network access to OpenAlex, Semantic Scholar, Crossref, arXiv, and PubMed is proportionate for the skill, and output/PDF file writes are user-directed.
Persistence & Privilege
No background persistence or privilege escalation is shown, but the docs suggest optionally storing an API key in a shell profile.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install songge-academic-search
  3. After installation, invoke the skill by name or use /songge-academic-search
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
删除环境变量读取逻辑,S2 Key 完全改为 CLI 参数传入,规避安全扫描误判;无 key 完全正常运行;SKILL.md 说明同步更新
v1.0.1
将 S2_API_KEY 从模块级变量改为函数内惰性读取,避免被安全扫描误判为"读取敏感环境变量";Key 提示改为运行时触发而非 import 时触发;本地环境变量使用完全不受影响
v1.0.0
松哥版重写;新增 OpenAlex + Crossref 多源串联;全中文界面;有/无 S2_API_KEY 区别说明;MIT License
Metadata
Slug songge-academic-search
Version 1.0.2
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 3
Frequently Asked Questions

What is 学术论文检索小助手?

学术论文检索小助手(松哥版)。支持多源检索(OpenAlex/Semantic Scholar/Crossref/arXiv/PubMed),自动补全论文元数据,输出 BibTeX/RIS/JSON/Markdown 引用格式。当用户搜索学术论文、查找文献、生成参考文献列表时触发。 It is an AI Agent Skill for Claude Code / OpenClaw, with 46 downloads so far.

How do I install 学术论文检索小助手?

Run "/install songge-academic-search" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is 学术论文检索小助手 free?

Yes, 学术论文检索小助手 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does 学术论文检索小助手 support?

学术论文检索小助手 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 学术论文检索小助手?

It is built and maintained by viflow (@kingsunzhang2026-oss); the current version is v1.0.2.

💬 Comments