BrunoSan ArXiv Intelligence

de.brunosan由 brunosan.de 发布官方发布者v1.0.0更新于 Oct 24, 2026

AI research intelligence for AI, ML, NLP, vision and robotics with citations and entity graphs.

已验证Streamable HTTP可网页运行网页搜索与抓取知识与记忆数据与分析

概览

AI 生成的概览

让助手检索实时 arXiv AI/ML 论文图谱，并查看引用关系、实体、作者、机构和关联代码仓库。

可以做什么
通过远程 arXiv 论文图谱提供 16 个工具，覆盖 cs.AI、机器学习、自然语言处理、视觉和机器人方向。助手可执行支持布尔运算符和短语匹配的全文检索，获取论文及其作者、参考文献和关联 GitHub 仓库，并对实体、作者和机构进行排名。还可追踪引用网络、查找同时提及两个实体的论文、按研究主题查看时间趋势，以及查询数据管道状态。
适用场景
适合文献综述、观察某个基准、模型或方法随时间兴起或衰退、按被引次数寻找有影响力的论文，以及了解哪些实验室和 GitHub 组织产出研究代码。
运行要求
通过 streamable HTTP 访问的远程端点；未声明任何软件包、环境变量、请求头或身份验证。需要能访问该端点的网络。
安装前请注意
数据来自第三方实时论文图谱，覆盖范围取决于其管道进度和回填进度，被引次数只统计数据库中具有可解析 arXiv ID 的论文。不要求任何凭据，也未描述写入、发送或删除类操作。

安装

在 SourceWeft 中

  1. 打开 控制台中的 BrunoSan ArXiv Intelligence，将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Web executable，通过 Streamable HTTP。 远程服务在工作区中配置后即可从网页运行时运行。

交给智能体安装

把这段提示词粘贴到 Claude Code、Codex、Cursor 等智能体，或 SourceWeft 对话中。智能体会读取该服务器的安装指南，先向你展示发布者、端点和登录方式，经你同意后再把它添加到你的客户端。密钥不会写进任何命令。

阅读 https://sourceweft.com/mcp/de.brunosan%2Farxiv/install.md，并按其中说明把这个 MCP 服务器接入我的客户端。修改任何配置之前，先告诉我它的发布者、端点和登录方式，等我确认。不要把 API 密钥写进命令或我会提交的文件；问我密钥应从哪里获取。

查看智能体所依据的安装指南

其他 MCP 客户端

把它添加到你客户端的 mcpServers 配置中。

{
  "mcpServers": {
    "brunosan-arxiv-intelligence": {
      "type": "http",
      "url": "https://arxiv.mcp.brunosan.de/mcp"
    }
  }
}

工具

16

arxiv_search_papers

Full-text search over the live cs.AI/ML paper graph using FTS5. Searches title AND abstract. Supports boolean operators: AND, OR, NOT, phrase matching ("exact phrase"), prefix (term*). Args: query: FTS5 search query. E.g. 'LoRA fine-tuning', '"chain of thought"', 'RLHF NOT PPO' category: Filter by primary category. Options: cs.AI, cs.LG, cs.CL, cs.CV, cs.RO date_from: ISO date filter, e.g. '2024-01-01' date_to: ISO date filter, e.g. '2025-12-31' empirical_only: Only papers marked as empirical by LLM pass (if available) has_code_only: Only papers with code release (llm_has_code=1, if available) limit: Max results (default: 20, max: 50)

输入参数limitquery必填api_keydate_tocategorydate_fromhas_code_onlyempirical_only

arxiv_get_paper

Full paper object with all connected data. Returns: paper metadata, author list with positions, matched entities, references (up to 100), and linked GitHub repos. Args: arxiv_id: ArXiv ID, e.g. '2402.01234' or '2402.01234v2'

输入参数api_keyarxiv_id必填

arxiv_top_entities

Entity ranking by mention count across all papers. title_only=True is a powerful relevance filter: a paper mentioning MMLU in the title IS about MMLU, not just using it as one of many benchmarks. Args: type: Filter by entity type: benchmark, model, method, dataset (optional, default: all) date_from: Only count mentions in papers published from this date date_to: Only count mentions in papers published until this date title_only: Only count mentions where entity appears in the paper title limit: Max results (default: 20, max: 50)

输入参数typelimitapi_keydate_todate_fromtitle_only

arxiv_entity_trend

How often is an entity mentioned over time? Shows the rise (or fall) of a benchmark, model, method, or dataset across the research literature — per month, quarter, or year. Example: 'LoRA' — watch it explode in 2023-2024. Example: 'BERT' — watch it decline as LLMs dominate. Args: entity_name: Entity to track, e.g. 'LoRA', 'MMLU', 'RAG', 'GPT-4' granularity: Time grouping: month (default), quarter, year

输入参数api_keyentity_name必填granularity

arxiv_top_authors

Top researchers ranked by paper count, with role filter. role='last_author' is the PI filter — finds lab directors and group leaders who drive research agendas. In academic AI, the last author IS the boss. role='first_author' finds the PhD students and postdocs doing the work. role='any' counts all papers regardless of position. Args: role: Author position filter: any (default), first_author, last_author category: Filter by primary category: cs.AI, cs.LG, cs.CL, cs.CV, cs.RO date_from: ISO date filter, e.g. '2024-01-01' date_to: ISO date filter, e.g. '2025-12-31' limit: Max results (default: 20, max: 50)

输入参数rolelimitapi_keydate_tocategorydate_from

arxiv_author_papers

All papers by a researcher, with their position on each paper. Uses fuzzy name matching (LIKE) to handle name variations. Returns papers sorted newest first. Args: author_name: Researcher name, e.g. 'Yann LeCun', 'lecun' (partial match works) limit: Max results (default: 20, max: 50)

输入参数limitapi_keyauthor_name必填

arxiv_most_cited

Most cited papers — ranked by inbound citation count. This answers the question every researcher, VC, and journalist asks first: 'What are the most influential papers in AI right now?' Counts how many papers in our database cite each target paper. Only papers with resolvable ArXiv IDs in their references are counted. Args: category: Filter citing papers by category (optional) date_from: Only count citations from papers published from this date date_to: Only count citations from papers published until this date limit: Max results (default: 20, max: 50)

输入参数limitapi_keydate_tocategorydate_from

arxiv_citation_network

Citation graph for a paper — who cites it, or what does it cite? direction='cited_by': Papers in our database that cite this paper. direction='citing': Papers that this paper cites (its references). depth=2: Expands one hop further (depth-2 neighbors). Hard cap: 200 total. Args: arxiv_id: ArXiv paper ID, e.g. '2402.01234' direction: 'cited_by' (inbound) or 'citing' (outbound, default: cited_by) depth: Graph depth: 1 or 2 (default: 1)

输入参数depthapi_keyarxiv_id必填direction

arxiv_co_occurrence

Papers that mention BOTH entity A and entity B. Answers questions like: - 'Which papers use both GPT-4 and RLHF?' - 'Where do LoRA and MMLU appear together?' - 'Papers combining RAG and Chain-of-Thought?' The intersection reveals research that explicitly bridges two concepts. Args: entity_a: First entity name, e.g. 'GPT-4', 'LoRA', 'MMLU' entity_b: Second entity name, e.g. 'RLHF', 'Chain-of-Thought' date_from: ISO date filter date_to: ISO date filter limit: Max results (default: 20, max: 50)

输入参数limitapi_keydate_toentity_a必填entity_b必填date_from

arxiv_institution_ranking

Institution ranking by paper count. Primary signal: author_affiliations extracted from ArXiv HTML. Secondary signal (include_github_orgs=True): adds GitHub org counts as a complementary signal. Many papers have no affiliation in HTML but do have a GitHub org link — combining both gives a fuller picture. Note: affiliation data is extracted from HTML and may be incomplete (fetch completion is reported live by arxiv_pipeline_status; extraction quality is a separate signal). Args: date_from: ISO date filter date_to: ISO date filter include_github_orgs: Also show GitHub org ranking as second signal limit: Max results (default: 20, max: 50)

输入参数limitapi_keydate_todate_frominclude_github_orgs

arxiv_repo_landscape

GitHub repository landscape — which orgs and repos produce research code? Shows the open-source output of the research community. 'openai', 'google-deepmind', 'microsoft', 'huggingface' etc. ranked by how many papers link to their repos. org_filter='huggingface' shows all HuggingFace repos with papers. Args: org_filter: Filter to a specific GitHub org, e.g. 'openai', 'google-deepmind' date_from: Only papers published from this date date_to: Only papers published until this date limit: Max results per ranking (default: 20, max: 50)

输入参数limitapi_keydate_todate_fromorg_filter

arxiv_tracks

List stable ArXiv research-track objects with definitions and live counts.

输入参数api_key

arxiv_track_papers

Chronological papers inside one stable research track. Args: track: Track slug or exact name, e.g. 'ai-agents' or 'multi-agent-systems'. date_from: Optional ISO date lower bound. date_to: Optional ISO date upper bound. has_code_only: Restrict to papers with confirmed code signal. limit: Max results (default 20, max 50).

输入参数limittrack必填api_keydate_todate_fromhas_code_only

arxiv_track_trend

Research volume for one stable track by month, quarter or year.

输入参数track必填api_keygranularity

arxiv_pipeline_status

Full system status — database counts, pipeline progress, frontier, quality. Returns: - Paper/author/entity/ref/repo counts - Pipeline progress: html_fetched %, whitelist_matched %, llm_processed % - Frontier: how far back the backfill has reached - Quality report: last run timestamp and overall status - Quality log: last 5 quality check runs from quality_log table

输入参数api_key

版本历史

1
  1. v1.0.0最新Sep 16, 2026

评分

已满足适用检查项的 72%

该评分衡量此收录条目的完整程度，依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价，也不表示服务器安全或当前可用：可连通性是带日期的检查结果，评分也不包含安装量或用户评价。

  • 近 30 天内可连通必需25 分
  • 已声明认证方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 图标10 分

    请在注册中心条目中发布图标。

  • 文档15 分

    请声明包含 README 的 GitHub 仓库，或在 OAuth 受保护资源元数据中提供 resource_documentation。

  • 许可证不适用

    托管服务按其提供方的条款使用，因此这里不要求许可证。此检查项不计入百分比。

其他 Brunosan 集成

BrunoSan Intelligence Concierge

de.brunosan官方发布者

远程发现服务器，让助手浏览、搜索并规划 BrunoSan 11 个情报领域加 LENS 网络中的路由。

Streamable HTTP可网页运行
效率与工作流9 个工具Oct 27, 2026