firecrawl-research-papers

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Firecrawl Research Papers

Firecrawl 研究论文工具

Use this to create a sourced literature review.

使用本工具创建带来源标注的文献综述。

Onboarding Interview

初始访谈

Infer the topic, source constraints, target count, and output format from context. If the topic is clear, proceed immediately.

Ask at most 1-3 concise questions only if blocked, such as the topic, target paper count, or required venue/date/method constraints.

从上下文推断研究主题、来源限制、目标论文数量和输出格式。若主题明确，可直接开展工作。

仅在遇到阻碍时提出最多1-3个简洁问题，例如询问研究主题、目标论文数量，或所需的发表平台/时间/研究方法限制。

Firecrawl Collection Plan

Firecrawl 文献收集方案

Use Firecrawl Research through the CLI, MCP, or equivalent Firecrawl tool surface as the primary path for paper discovery and verification. Fall back to general Firecrawl search and scrape for whitepapers, technical reports, research blogs, leaderboards, or facts outside the paper corpus.

What the paper index holds: paper abstracts, with full text reachable per paper. Its largest share is biomedical and life-science literature — PubMed journal articles plus bioRxiv and medRxiv preprints — so clinical, drug, gene, disease, epidemiology, and public-health questions are in scope. arXiv preprints cover computer science, physics, and mathematics. Coverage outside those sources is thinner, and the web tools below are the fallback there.

Core tools:

MCP:
```
firecrawl_research_search_papers(query, k?)
```
CLI:
```
firecrawl research search-papers <query> [--k <number>]
```
Semantic search over paper abstracts. Start here for most paper-finding queries, and retry with alternate framing when results are thin or too narrow.

MCP:

firecrawl_research_related_papers(seed_ids, intent, mode?, k?)

CLI:

firecrawl research related-papers <seedIds...> --intent <intent> [--mode <similar|citers|references>] [--k <number>]

Expand from strong seed papers into similar work, citing papers, or references. Use this to find the relevant paper family, not just the first matching result.

MCP:
```
firecrawl_research_inspect_paper(id)
```
CLI:
```
firecrawl research inspect-paper <id>
```
Fetch canonical metadata for a candidate paper: title, abstract, authors, categories, source ids, and dates.
MCP:
```
firecrawl_research_read_paper(id, question)
```
CLI:
```
firecrawl research read-paper <id> --question <question>
```
Verify a specific claim or constraint inside one paper, such as method, reported score, benchmark, affiliation, comparison, or limitation.
MCP:
```
firecrawl_search(query)
```
/
```
firecrawl_scrape(url)
```
CLI:
```
firecrawl search <query>
```
/
```
firecrawl scrape <url>
```
Use for web-only context: benchmark leaderboards, rankings, reports, whitepapers, research blogs, and source pages outside the paper index.

Not the paper index, despite the name: passing

categories: ["research"]

firecrawl_search

(CLI

firecrawl search <query> --categories research

) filters an ordinary web search to research-affiliated websites — the list includes PubMed, bioRxiv, medRxiv, arXiv, and publisher sites — and returns page results from them. It reaches those sites' web pages; what it does not do is query their paper records in the index above, so there is no abstract search, no related-paper or citation-graph expansion, no canonical paper metadata, and no in-body passages. Use it when a web search is what you want and those sites should be weighted in the same call; use the

firecrawl_research_*

tools for paper work.

Match the approach to the query:

Single named paper: run one paper search, then inspect or read the paper if metadata or body verification is needed.
Paper by description, method, or topic family: search for strong anchors, then expand with related papers and keep close neighbors.
Enumeration queries, such as papers that do a task or benchmark a method: search multiple framings, expand several strong anchors, and re-seed from newly found relevant papers.
Papers that use or exhibit a property: start from the defining paper or strongest anchor, expand via similar, citers, or references, and use read-paper to verify the property.
Superlatives and leaderboards: use general web search or scrape to find the ranking, then map top entries back to papers with paper search.
Author, organization, venue, date, or methodology constraints: verify with inspect-paper metadata or read-paper before keeping a candidate.

Target source types:

biomedical and life-science literature from PubMed, with bioRxiv and medRxiv preprints for work that has not appeared in a journal yet
arXiv preprints in computer science, physics, and mathematics
academic papers from university sites and ACM/IEEE pages where accessible
industry reports and whitepapers
company research blogs
technical articles and conference summaries

Principles:

When in doubt, include the relevant paper family rather than only the single best result.
Use related-paper expansion to avoid stopping at one strong hit.
Use read-paper to verify load-bearing constraints, not to summarize every candidate.
Drop only clearly off-topic papers.

通过CLI、MCP或等效Firecrawl工具调用Firecrawl Research，将其作为论文发现与验证的主要途径。若需查找白皮书、技术报告、研究博客、排行榜或论文索引以外的信息，则退而使用通用Firecrawl搜索与爬取功能。

论文索引包含的内容：论文摘要，且可获取单篇论文的全文。其中占比最大的是生物医学与生命科学文献——PubMed期刊文章及bioRxiv、medRxiv预印本，因此临床、药物、基因、疾病、流行病学和公共卫生领域的问题均在适用范围内。arXiv预印本涵盖计算机科学、物理学和数学领域。上述来源以外的内容覆盖范围较窄，此时需使用下方的网页工具作为备选方案。

核心工具：

MCP：
```
firecrawl_research_search_papers(query, k?)
```
CLI：
```
firecrawl research search-papers <query> [--k <number>]
```
基于论文摘要进行语义搜索。大多数论文查找查询都从这里开始，若结果数量少或范围过窄，可尝试换一种表述方式重试。
MCP：
```
firecrawl_research_related_papers(seed_ids, intent, mode?, k?)
```
CLI：
```
firecrawl research related-papers <seedIds...> --intent <intent> [--mode <similar|citers|references>] [--k <number>]
```
从优质种子论文扩展到相似研究、引用该论文的文献或该论文的参考文献。使用此工具可找到相关的论文集群，而非仅局限于首个匹配结果。
MCP：
```
firecrawl_research_inspect_paper(id)
```
CLI：
```
firecrawl research inspect-paper <id>
```
获取候选论文的标准元数据：标题、摘要、作者、分类、来源ID和发表日期。
MCP：
```
firecrawl_research_read_paper(id, question)
```
CLI：
```
firecrawl research read-paper <id> --question <question>
```
验证单篇论文中的特定声明或限制条件，例如研究方法、报告分数、基准测试、机构归属、对比分析或局限性。
MCP：
```
firecrawl_search(query)
```
/
```
firecrawl_scrape(url)
```
CLI：
```
firecrawl search <query>
```
/
```
firecrawl scrape <url>
```
用于仅存在于网页中的内容：基准测试排行榜、排名、报告、白皮书、研究博客，以及论文索引以外的来源页面。

注意：尽管名称相似，但向

firecrawl_search

传入

categories: ["research"]

（CLI命令为

firecrawl search <query> --categories research

）只是将普通网页搜索过滤为学术机构相关网站——包括PubMed、bioRxiv、medRxiv、arXiv和出版商网站——并返回这些网站的页面结果。它仅能访问这些网站的网页，无法查询上述论文索引中的论文记录，因此不支持摘要搜索、相关论文或引用图谱扩展、标准论文元数据获取以及正文段落提取。当你需要进行网页搜索，且希望在同一调用中优先考虑这些网站时使用此命令；若处理论文相关工作，请使用

firecrawl_research_*

系列工具。

根据查询类型匹配对应方法：

查找特定命名的论文：执行一次论文搜索，若需要元数据或正文验证，则调用inspect或read-paper工具。
根据描述、方法或主题类别查找论文：先搜索优质锚点论文，再通过相关论文扩展功能获取相近的研究成果。
枚举类查询（例如查找完成某项任务或测试某方法的论文）：使用多种表述方式进行搜索，扩展多个优质锚点论文，并从新找到的相关论文中重新选取锚点。
查找具备特定属性的论文：从定义性论文或最优质的锚点论文开始，通过相似论文、引用论文或参考文献进行扩展，并使用read-paper工具验证该属性。
最高级类查询和排行榜：使用通用网页搜索或爬取功能查找排名，再通过论文搜索将排名靠前的条目对应到具体论文。
带有作者、机构、发表平台、时间或研究方法限制的查询：在保留候选论文前，使用inspect-paper元数据或read-paper工具进行验证。

目标来源类型：

PubMed的生物医学与生命科学文献，以及尚未发表在期刊上的bioRxiv和medRxiv预印本
arXiv中计算机科学、物理学和数学领域的预印本
可访问的大学网站及ACM/IEEE页面中的学术论文
行业报告和白皮书
企业研究博客
技术文章和会议摘要

原则：

若存在疑问，应纳入相关的论文集群，而非仅保留单一最优结果。
使用相关论文扩展功能，避免仅停留在首个优质结果。
使用read-paper工具验证关键限制条件，而非对每个候选论文进行摘要。
仅剔除明显偏离主题的论文。

Parallel Work

并行工作

If appropriate, use sub-agents or equivalent parallel task runners:

Academic Papers researcher
Biomedical and Life Sciences researcher, for PubMed journal articles and bioRxiv/medRxiv preprints on a clinical, drug, gene, disease, epidemiology, or public-health topic
Industry Reports researcher
Technical Articles researcher
Synthesis and citation reviewer

Split by source or sub-topic, not by tool. Give each researcher the same paper tools and let the topic decide which part of the corpus answers.

若合适，可使用子Agent或等效的并行任务运行器：

学术论文研究员
生物医学与生命科学研究员（负责处理PubMed期刊文章及bioRxiv/medRxiv中关于临床、药物、基因、疾病、流行病学或公共卫生主题的预印本）
行业报告研究员
技术文章研究员
综述与引用审核员

按来源或子主题划分任务，而非按工具划分。为每位研究员提供相同的论文工具，由研究主题决定使用语料库的哪部分内容来回答问题。

Final Deliverable

最终交付物

markdown

undefined

markdown

undefined

Literature Review: [Topic]

文献综述：[研究主题]

Abstract

摘要

[2-3 paragraph summary]

[2-3段总结]

Key Papers

关键论文

[Title, authors, source URL, key findings, methodology, relevance]

[标题、作者、来源URL、核心发现、研究方法、相关性]

Themes And Consensus

主题与共识

[What sources agree on]

[各来源达成的共识内容]

Open Questions And Debates

开放问题与争议

[Disagreements and unresolved questions]

[存在分歧和未解决的问题]

Emerging Trends

新兴趋势

[Recent developments]

[近期发展动态]

Sources

来源

[Organized by paper/report/article]

[按论文/报告/文章分类整理]

Rerun Inputs

重运行参数

workflow: firecrawl-research-papers topic: [topic] target_count: [number] output: [markdown/brief]

undefined

workflow: firecrawl-research-papers topic: [研究主题] target_count: [数量] output: [markdown/精简版]

undefined

Quality Bar

质量标准

Every major claim should trace to a source.
Note inaccessible or failed PDFs.
Distinguish peer-reviewed work from blogs and vendor reports.

每个重要声明都应可追溯到对应来源。
标注无法访问或加载失败的PDF文件。
区分同行评审成果与博客、供应商报告。",