researching-academic-literature

name	researching-academic-literature
description	Use when a user needs real, high-relevance academic literature for a research topic, manuscript section, experiment, model, theory, review, method comparison, discussion, or reference-list strengthening.

Researching Academic Literature

Overview

Find, verify, screen, and organize real academic literature that can directly support scholarly writing. Treat the user's research topic as an input; never assume a fixed discipline or process.

Required Inputs

1. Identify the user's literature-search topic and intended use, such as a review, experiment, model, method, theory, introduction, discussion, or comparison.
2. If the topic is missing or too broad to search responsibly, ask for a focused scientific question before recommending papers.
3. When a manuscript draft, project file, or reference list is available and relevant, read it first to align the search with the paper's actual scope. Do not modify the manuscript unless explicitly asked.
4. Extract English keywords, synonyms, abbreviations, related processes, and adjacent terminology before searching.

Search And Verification Workflow

1. Search reliable scholarly sources, including DOI records, Crossref, publisher or journal pages, Google Scholar, PubMed, Web of Science, and other trusted databases appropriate to the field.
2. Prioritize papers that are directly relevant, highly citable, methodologically useful, and published in strong journals.
3. Prefer, but do not limit the search to, these journals:
   • Geochimica et Cosmochimica Acta
   • Environmental Science & Technology
   • Journal of Geophysical Research: Solid Earth
   • Water Research
   • Earth and Planetary Science Letters
   • Geophysical Research Letters
   • Nature Communications
   • Advanced Materials
   • Advanced Functional Materials
   • Proceedings of the National Academy of Sciences of the United States of America
   • Science Advances
   • Physical Review Letters
   • Nature
   • Science
   • Nature Geoscience
4. Include papers from other journals when they are highly relevant, and explain why they merit inclusion.
5. Verify every recommended paper before presenting it. Confirm as many of these fields as possible: title, authors, year, journal, DOI, and publisher page.
6. Never invent papers, authors, dates, journals, volume or issue data, pages, DOIs, or links. Label an item 未能核实 and exclude it from reliable recommendations when its existence or metadata cannot be confirmed.
7. Rank results by relevance to the user's scientific question, not merely by year, journal prestige, or search-engine order.
8. Explain why each paper is worth reading and how it can serve the user's work. Do not return a bare title list or pad the answer with low-relevance papers.

Evidence Language

• Distinguish among 文献明确证明, 文献支持, and 可以作为间接参考.
• Do not overstate a paper's conclusions.
• Mark inferences explicitly as interpretations based on the literature.
• Use precise academic terminology and retain English terms where useful.
• Answer in Chinese by default unless the user requests another language.

Default Output

Use this structure unless the user requests a different format:

### 检索主题

简要说明科学问题、检索目标、关键词和同义表达。

### 推荐文献

1. Title
   Authors, Year, Journal
   DOI / Link:
   相关性说明：
   适合用于：

### 总结建议

说明哪些文献适合作为核心引用、方法支撑、讨论对比或后续检索起点，
并指出当前证据链仍缺少的文献方向。

Adapt 适合用于 to the user's topic, such as introduction background, mechanism discussion, experimental design, model methods, analytical methods, theory, or comparison with the user's results.

Quality Check

Before responding, confirm:

• The search topic came from the user or their provided manuscript context.
• Every reliable recommendation has been individually verified.
• The list is ordered by scientific relevance.
• Each paper includes a specific relevance explanation and intended use.
• Unverified or weakly related items are clearly separated or omitted.