← Back to Skills Marketplace
agjvsxgm

心理学文献搜索与整理

by Agjvsxgm · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ⚠ suspicious
41
Downloads
4
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install psych-literature-search
Description
Searches and organizes academic literature on a given topic or research question. Queries PubMed (NCBI E-utilities, no API key required) and Semantic Scholar...
README (SKILL.md)

\r \r

文献自动搜索与整理 Skill\r

\r

功能目的\r

\r 针对用户给定的研究主题,自动搜索 PubMed(NCBI E-utilities,无需 API Key)与 Semantic Scholar(免费 REST API,无需 Key)两大数据库,提取每篇文献的核心信息,并整理为结构化表格,包含:文献标题、关键词、链接、摘要、研究变量的概念定义、操作性定义(测量工具)等。\r \r ---\r \r

触发条件(Trigger Conditions)\r

\r 当用户出现以下任一意图时,立即启用此 Skill:\r \r 中文触发词:\r

  • 帮我搜索 / 查找 / 找 XX 主题的文献\r
  • 做一个文献综述 / 文献检索\r
  • 整理 XX 相关的研究论文\r
  • 文献搜索,主题是 XX\r
  • 帮我找关于 XX 的学术文章\r
  • XX 的研究变量是什么 / 操作性定义是什么\r
  • 搜索 PubMed / Semantic Scholar 文献\r
  • 搜索心理学 / 医学 / 生命科学文献\r \r English trigger phrases:\r
  • "search literature on [topic]"\r
  • "find papers / articles about [topic]"\r
  • "literature review on [topic]"\r
  • "search PubMed / Semantic Scholar for [topic]"\r
  • "organize papers into a table"\r
  • "what are the research variables / operational definitions in studies on [topic]"\r \r ---\r \r

执行工作流(Workflow)\r

\r

Step 1 — 确认搜索参数\r

\r If the query is unclear, ask the user:\r

  • 研究主题(必填):具体关键词或研究问题\r
  • 时间范围(默认近 10 年)\r
  • 文献数量(默认每个数据库 10–15 篇)\r
  • API 密钥(可选):WoS / Springer 密钥;无密钥则使用 Semantic Scholar 免费备用通道\r \r 建议澄清提示(中文):\r

"请告诉我:① 具体研究主题或关键词;② 时间范围(默认近10年);③ 需要多少篇?如有 Web of Science 或 Springer API 密钥请提供,没有也可以用免费通道搜索。"\r \r ---\r \r

Step 2 — 搜索文献\r

\r PubMed 和 Semantic Scholar 均无需 API Key,直接运行附带脚本即可。\r \r

并行搜索:PubMed + Semantic Scholar(推荐)\r

\r 使用托管 Python 运行时并行运行两个脚本:\r \r PubMed(NCBI E-utilities,无需 Key):\r

"C:\Users\refresh\.workbuddy\binaries\python\versions\3.13.12\python.exe" \\r
  "C:/Users/refresh/.workbuddy/skills/psych-literature-search/scripts/search_pubmed.py" \\r
  --query "[TOPIC]" --max 30\r
```\r
\r
**Semantic Scholar(免费 REST API,无需 Key):**\r
```bash\r
"C:\Users\refresh\.workbuddy\binaries\python\versions\3.13.12\python.exe" \\r
  "C:/Users/refresh/.workbuddy/skills/psych-literature-search/scripts/search_semantic.py" \\r
  --query "[TOPIC]" --max 30\r
```\r
\r
两个脚本并行运行;解析 JSON 输出,每个返回 `hits` 数组,包含 title, authors, year, journal, doi, abstract, keywords, url, database 字段。\r
\r
**PubMed 参数说明:**\r
- `--mindate 2015` / `--maxdate 2024`:时间范围过滤\r
- `--email [email protected]`:可选,填写你的邮箱以符合 NCBI 礼仪要求\r
\r
**Semantic Scholar 参数说明:**\r
- `--year 2015-2024`:年份过滤\r
- `--venue "Journal Name"`:按期刊/会议过滤\r
- 速率限制:无需 Key 时 100 次/5 分钟,脚本已内置限速(每次请求间隔 6 秒)\r
\r
#### 方案 B:WebFetch 补充(可选)\r
\r
若脚本执行失败,使用 WebFetch 工具直接调用 API:\r
\r
**PubMed E-utilities(WebFetch):**\r
```\r
https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi?db=pubmed&term=[URL-ENCODED TOPIC]&retmax=20&sort=relevance&retmode=xml\r
```\r
然后对返回的每个 PMID 调用:\r
```\r
https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=pubmed&id=[PMIDs]&retmode=xml\r
```\r
\r
**Semantic Scholar(WebFetch):**\r
```\r
https://api.semanticscholar.org/graph/v1/paper/search?query=[URL-ENCODED TOPIC]&fields=title,authors,year,venue,doi,abstract,citationCount,fieldsOfStudy&limit=20\r
```\r
\r
#### 方案 C:WebSearch 兜底\r
\r
如脚本和 WebFetch 均失败,使用 WebSearch:\r
- `"[topic]" pubmed abstract`\r
- `"[topic]" site:semanticscholar.org`\r
- `"[topic]" (site:pubmed.ncbi.nlm.nih.gov OR site:semanticscholar.org)`\r
\r
---\r
\r
### Step 3 — 补充被引次数与文献类型\r
\r
PubMed 结果已包含 `publication_types`(如 "Journal Article", "Review", "Meta-Analysis")。\r
Semantic Scholar 结果已包含 `citations`(被引次数)和 `influential_citations`(高影响力被引次数)。\r
\r
无需额外脚本调用,直接使用 Step 2 返回的数据即可。\r
\r
**可选**:如需补充期刊水平信息(JCR 分区、影响因子),可运行(需要相应 API 访问权限):\r
```bash\r
"C:\Users\refresh\.workbuddy\binaries\python\versions\3.13.12\python.exe" \\r
  "C:/Users/refresh/.workbuddy/skills/psych-literature-search/scripts/journal_level.py" \\r
  --journal "[JOURNAL NAME]"\r
```\r
\r
未查询到分区信息时,注明"详见 JCR/Scimago"并附链接:\r
- https://jcr.clarivate.com/jcr/browse-journals\r
- https://www.scimagojr.com/\r
\r
---\r
\r
### Step 4 — 去重与排序\r
\r
1. 按 DOI 合并 PubMed 与 Semantic Scholar 重复记录(标注 `[双库收录]`)\r
2. 排序:① 被引次数降序 ② 发表年份降序\r
3. 保留用户要求数量(默认 15 篇)\r
\r
---\r
\r
### Step 5 — 生成结构化表格\r
\r
按照 `references/output_templates.md` 中的格式,输出以下内容:\r
\r
**主输出:一体化文献整理表(核心需求)**\r
\r
输出一张包含所有字段的综合表格,列包括:\r
\r
| 编号 | 标题(含链接) | 关键词 | 摘要(原文) | 摘要(中文) | 研究变量 | 概念定义 | 操作性定义 | 期刊水平 | 被引次数 |\r
|------|-------------|--------|-------------|------------|---------|---------|----------|---------|---------|\r
\r
若一篇文献有多个研究变量,将同一编号的行合并或拆分多行分别描述每个变量。\r
\r
**辅助四表结构(用于系统化综述报告):**\r
\r
- **表一** — 文献概览:标题、作者、年份、期刊、期刊水平、被引、DOI/链接\r
- **表二** — 关键词与摘要:关键词(双语)、摘要原文、摘要中文翻译\r
- **表三** — 研究变量定义表:研究变量、概念定义、操作性定义(测量工具/量表)\r
- **表四** — 研究设计摘要:研究类型、样本量、方法、主要工具、统计方法\r
\r
**综合分析部分**(可选,系统综述时使用):\r
- 核心发现(3-5 条,注明文献编号)\r
- 研究方法趋势\r
- 研究空白与未来方向\r
\r
**APA 7 参考文献列表**\r
\r
详见 `references/output_templates.md` 中的完整模板。\r
\r
---\r
\r
### Step 6 — 交付报告\r
\r
- 在对话中以 Markdown 形式展示表格\r
- 如用户需要文件,写入 `[主题]_文献整理_[日期].md` 到工作区\r
- 可按需提供 Word 兼容格式\r
\r
---\r
\r
## 关键参考文件\r
\r
- `references/api_reference.md` — PubMed E-utilities 与 Semantic Scholar API 端点、字段映射、速率限制\r
- `references/output_templates.md` — 报告模板与输出格式规则(含所有表格示例)\r
- `scripts/search_pubmed.py` — PubMed 搜索脚本(无需 API Key)\r
- `scripts/search_semantic.py` — Semantic Scholar 搜索脚本(无需 API Key)\r
- `scripts/journal_level.py` — 期刊影响因子/分区查询脚本(可选)\r
\r
---\r
\r
## 重要注意事项\r
\r
- **链接格式**:对每篇文献,优先使用 `https://doi.org/[DOI]` 作为标准链接;无 DOI 时使用 PubMed 链接或 Semantic Scholar 链接。\r
- **关键词处理**:若文献未显式列出关键词,从摘要和主题标签中推断 3–6 个关键词,用 `(推断)` 标注。双语格式:英文原词 + 中文翻译,如 `cognitive empathy(认知共情)`。\r
- **研究变量定义提取规则**:\r
  - 实证研究:列出所有被测量的核心变量(自变量、因变量、中间变量),分别给出概念定义(理论定义)和操作性定义(具体测量工具、量表名称、题目数量、计分方式)。\r
  - 综述/元分析:列出所综述的主要构念及其测量方式的范围。\r
  - 若文献未显式给出概念定义,从摘要和背景中推断,标注 `(推断)`。\r
  - 尽量包含信度信息(如 Cronbach's α)。\r
- **摘要翻译**:英文摘要提供完整中文翻译(非摘要摘要);超过 300 词时先压缩至约 200 词再翻译。\r
- **研究领域范围**:默认搜索覆盖所有学科,若用户明确心理学主题,在 PubMed 中可加 MeSH 限定词,在 Semantic Scholar 中加 `fieldsOfStudy:Psychology` 过滤。\r
- **摘要缺失处理**:标注 `[摘要不可获取,请查阅原文]`。\r
- **中文主题**:若主题为中文,先翻译为英文再检索,并在报告中注明翻译。\r
- **伦理约束**:不尝试下载或访问全文 PDF;仅通过官方 API 获取元数据和摘要。\r
- **API 密钥安全**:不在最终报告中打印或记录 API 密钥。\r
Usage Guidance
Install only if you are comfortable with public literature queries being sent to PubMed/Semantic Scholar and with AI translation/formatting of abstracts. Do not paste or store Web of Science, Springer, or other API keys unless the publisher clarifies exactly how those keys are used and updates the documentation to match the actual scripts.
Capability Tags
requires-sensitive-credentials
Capability Assessment
Purpose & Capability
The core purpose is coherent: search academic literature, call PubMed and Semantic Scholar, and format public metadata/abstracts. However, README and reference templates also advertise Web of Science/Springer searches, JCR/SSCI enrichment, and optional API keys while the manifest and bundled search scripts center on PubMed/Semantic Scholar.
Instruction Scope
Activation phrases are broad but aligned with literature-search intent. The stronger concern is scope drift: runtime guidance asks for optional WoS/Springer keys and output templates label sources as Web of Science/Springer even though those integrations are not clearly implemented.
Install Mechanism
The README describes ZIP import or cloning into a WorkBuddy skills directory. No install-time executable behavior, package installation, obfuscation, or destructive setup was found.
Credentials
The bundled scripts make network requests to PubMed and Semantic Scholar and print JSON results, which is proportionate for literature search. Optional guidance about storing API keys and using undeclared providers is under-scoped for the advertised no-key design.
Persistence & Privilege
No background persistence, privilege escalation, credential harvesting, or broad local indexing was found. File output is described only when the user asks for a report file, but users should confirm output paths and avoid storing secrets in skill files.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install psych-literature-search
  3. After installation, invoke the skill by name or use /psych-literature-search
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
Version 1.0.1 — Switched to PubMed and Semantic Scholar as primary data sources. Streamlined, API-key-free workflow. - Replaced Web of Science and Springer with PubMed (NCBI E-utilities) and Semantic Scholar as main sources; both require no API key. - Removed dependency on search scripts using proprietary APIs (search_wos.py, search_springer.py deleted). - Added dedicated scripts for PubMed and Semantic Scholar queries. - Updated documentation and workflows to reflect new data sources, processes, and field mappings. - Now outputs citation counts and publication types in structured tables. - All search, extraction, and reporting processes now work without requiring user API credentials.
v1.0.0
- Initial release of psych-literature-search skill. - Enables automated search and organization of academic literature via Web of Science and Springer Nature APIs (with Semantic Scholar as fallback). - Generates structured tables with bilingual keywords, abstracts (with Chinese translation), research variable definitions, operational definitions, journal quality metrics, and APA 7 citations. - Supports trigger phrases in both Chinese and English for literature search, review, and variable extraction tasks. - Provides detailed workflow for parameter confirmation, search execution, journal ranking retrieval, deduplication, and report delivery in Markdown format.
Metadata
Slug psych-literature-search
Version 1.0.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is 心理学文献搜索与整理?

Searches and organizes academic literature on a given topic or research question. Queries PubMed (NCBI E-utilities, no API key required) and Semantic Scholar... It is an AI Agent Skill for Claude Code / OpenClaw, with 41 downloads so far.

How do I install 心理学文献搜索与整理?

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

Is 心理学文献搜索与整理 free?

Yes, 心理学文献搜索与整理 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does 心理学文献搜索与整理 support?

心理学文献搜索与整理 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 心理学文献搜索与整理?

It is built and maintained by Agjvsxgm (@agjvsxgm); the current version is v1.0.1.

💬 Comments