Reviewmetric

io.github.vktr93由 GitHub 账号 vktr93 发布v1.0.6更新于 Oct 24, 2026

An MCP server connecting Claude to the Review Metric 15-agent scientific peer review pipeline.

已验证STDIO仅桌面知识与记忆AI 与机器学习

概览

AI 生成的概览

让助手把本地 .docx 稿件提交到 Review Metric 多智能体同行评审流程，并返回质量评分。

可以做什么
该 MCP 服务器把助手连接到 Review Metric 科学同行评审服务。它可以读取本机上的 .docx 稿件，并提交给由 15 个智能体组成的流程，分析方法学缺陷、新颖性和文献契合度。结果包含 0-100 的质量评分，用于判断是否适合投稿期刊。
适用场景
适合研究人员、学者和科学家，希望让助手自动完成稿件评审，而不必把长文档粘贴到对话中。适用于投稿前检查和草稿的早期反馈。
运行要求
作为 PyPI 包（reviewmetric-mcp）在本地运行，通过 uvx 启动，因此需要 Python 工具链。需要 Review Metric API 密钥，通过环境变量 REVIEWMETRIC_API_KEY 提供，密钥来自 Review Metric 网页控制台。仅支持桌面端；需要访问 Review Metric 服务的网络连接。
安装前请注意
该服务器会读取本地 .docx 稿件文件并将其内容发送到第三方服务，因此未发表的作品会离开本机。它需要密钥 REVIEWMETRIC_API_KEY，应避免写入共享配置文件。README 声称零数据保留且研究内容不会用于 AI 训练，但这只是提供方自己的说法，应自行核实。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Reviewmetric，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。

Desktop only，通过 STDIO。 STDIO 服务会启动本地进程，因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

ReviewMetric MCP Server

Know what reviewers will say before you submit — from inside Claude.
Then fix the text in one go, and see where your work sits on the map of science.

ReviewMetric is a pre-submission review service for researchers. This MCP server lets Claude send a local .docx manuscript to ReviewMetric's panel of specialist AI referees, bring the report back into your chat, prepare the text fixes Claude can apply to your file, and explore the ReviewMetric Atlas: landmark discoveries, their Nobel Prizes and the frontiers research groups are pushing now.

What the panel checks

RefereeWhat it looks at
Novelty & positioningYour claims against the closest published, peer-reviewed work (retrieved from OpenAlex and ReviewMetric's vetted library), plus related papers you may have missed
Claims & evidenceWhether each main claim is supported by your results; numbers that disagree between abstract, results and tables
Methods & reproducibilityDesign, controls, statistics, and a field-appropriate reproducibility checklist
Citation integrityWhether cited papers support the sentences they are attached to; retracted, uncited or missing references
FiguresCaptions, legibility, units, and agreement with the text
ClarityTitle, abstract, structure, language, terminology
Handling editorReadiness recommendation, prioritised revisions and a referee-style report

Every issue points to a location in your manuscript. Scores come from fixed rubrics, not from a single AI guess.

Installation (Claude Desktop)

  1. Create an API key at https://www.reviewmetric.it.com/developers
  2. Add this to claude_desktop_config.json:
json
{  "mcpServers": {    "review-metric": {      "command": "uvx",      "args": ["reviewmetric-mcp"],      "env": { "REVIEWMETRIC_API_KEY": "rvk_live_..." }    }  }}
  1. Ask Claude: "Review my manuscript at /Users/me/papers/draft.docx with ReviewMetric."

Commands

Ask "What can ReviewMetric do?" (the help_me command) for the full list inside Claude.

CommandWhat it doesCost
help_meEverything ReviewMetric can do in Claude, with examplesFree
submit_manuscript(file_path)Sends the .docx to the referees; returns a Run ID1 RM token
check_analysis_status(run_id)Progress; when done, the report, its place on the Atlas and the text fixes availableFree
get_report_json(run_id)The complete report as JSONFree
list_my_reviews(limit)Your reviews with scores and Run IDsFree
get_text_fixes(run_id)Preview of the fixes that need no new experiment; the exact edits once preparedFree
prepare_text_fixes(run_id)Prepares the edits, which Claude can then apply to your manuscript1 RM token
search_atlas(query, field, limit)Finds discoveries and frontiers on the AtlasFree
list_frontiers()The frontiers, what they build on, who is pushing themFree
place_on_atlas(run_id)Where your manuscript sits, the landmarks it cites, the frontier aheadFree
atlas_whats_new()This week's additions to the AtlasFree
trace_path(from, to)The chain of discoveries linking two starsFree
unlock_atlas_sources(confirm)The paper and research group behind every star, for 30 days1 RM token
account_status()RM tokens left, Atlas access, pricesFree

Spending tokens. RM tokens are ReviewMetric's one currency (1 for $9, 3 for $25, 10 for $70; your first is free). Claude never unlocks the Atlas sources on its own: it tells you the cost and waits for your explicit yes in the chat.

A review usually takes about 5 minutes. If the file cannot be reviewed (unreadable, too short, not a manuscript), the RM token is returned automatically.

Try asking

  • "Review my manuscript at ~/papers/draft.docx and tell me what to fix first."
  • "Prepare the text fixes of my last review and apply them to the draft."
  • "Where does my manuscript sit on the Atlas, and which frontier is ahead of it?"
  • "Who is pushing room-temperature superconductivity right now?"
  • "Trace the path from the double helix to AlphaFold."

Data

Manuscripts are processed by ReviewMetric's API and the AI providers it uses under their paid API terms (no training on your data). See the ReviewMetric website for the current data-retention policy.

来源：README.md，提交 b7167c6

工具

0
SourceWeft 尚未在沙箱中启动过该软件包，因此其工具未知。

版本历史

1
  1. v1.0.6最新Sep 16, 2026

评分

已满足适用检查项的 50%

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25 分

    请保持端点能够响应 MCP 握手；SourceWeft 每 30 天探测一次。

  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头，或保持端点能够响应探测，以便判定其认证方式。

  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list，或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分