Skip to main content
Glama
iapke486-arch

paper-organizer-mcp

by iapke486-arch
OverviewSchemaSourceRelated ServersScoreDiscussionsAdmin
TypeScript
Local

paper-organizer-mcp

로컬 폴더에 있는 논문 PDF를 스캔하고, 제목/초록 텍스트를 추출해 분류 판단(Claude 또는 다른 분류 MCP)에 활용할 수 있도록 하며, 결정된 분류에 따라 지정한 폴더 구조로 파일을 이동/정리하는 MCP(Model Context Protocol) 서버입니다.

이 서버는 분류 로직 자체는 포함하지 않습니다. "이 논문이 어떤 카테고리에 속하는지"는 Claude가 paper_read_text로 얻은 텍스트를 보고 직접 판단하거나, 별도의 분류 전용 MCP에 위임하는 것을 전제로 합니다. 이 서버는 스캔 → 텍스트 확인 → 이동/정리, 세 단계만 담당합니다.

설치 및 빌드

git clone https://github.com/iapke486-arch/paper-organizer-mcp.git
cd paper-organizer-mcp

npm install
npm run build

Related MCP server: renfield-mcp-filesystem

실행 (직접 테스트)

node build/index.js

Claude Desktop / Claude Code 설정

Claude Desktop

claude_desktop_config.json(보통 %APPDATA%\Claude\claude_desktop_config.json)에 등록합니다.

{
  "mcpServers": {
    "paper-organizer": {
      "type": "stdio",
      "command": "node",
      "args": ["C:/Users/YourName/Documents/pke/paper-organizer-mcp/build/index.js"]
    }
  }
}

경로 예시:

  • Windows: C:/Users/YourName/Documents/pke/paper-organizer-mcp/build/index.js

  • macOS/Linux: /home/username/projects/paper-organizer-mcp/build/index.js

Claude Code CLI

claude mcp add paper-organizer -- node /path/to/paper-organizer-mcp/build/index.js

연결 확인

Claude Code: /mcp 목록에서 paper-organizer 서버가 ✓ Connected 상태인지 확인하세요.

제공 도구 (Tools)

Tool

설명

paper_scan_folder

지정한 폴더(재귀 옵션 지원)에서 PDF 파일 목록을 스캔. 각 파일의 절대 경로/크기/수정일 반환

paper_read_text

PDF 앞부분 페이지(기본 3페이지)의 텍스트와 메타데이터(Author 등)를 추출해 분류 판단 근거로 사용

paper_move

분류가 결정된 PDF를 destRoot/category 경로로 이동. 폴더가 없으면 자동 생성, 필요 시 fileName으로 이름 변경

사용 흐름 예시

  1. "C:/papers/inbox 폴더에 있는 논문들을 분류해서 C:/papers/library 밑에 정리해줘" 라고 요청

  2. Claude가 paper_scan_folder로 inbox의 PDF 목록을 확인

  3. 각 파일에 대해 paper_read_text로 제목/초록을 읽고 분류 카테고리 판단 (또는 별도 분류 MCP 결과 활용)

  4. paper_move로 destRoot=C:/papers/library, category="NLP" 등과 같이 원하는 폴더 구조로 이동

폴더 경로(destRoot)와 분류 체계(category)는 고정되어 있지 않으며, 매 요청마다 자유롭게 지정할 수 있습니다.

참고

  • 논문 파일이 이미 다른 위치로 이동되면 원래 폴더의 스캔 결과에서 자연히 제외되므로, "처리 완료" 여부를 별도로 기록/추적하지 않습니다.

  • paper_move는 대상 경로에 동일한 이름의 파일이 있으면 덮어쓰지 않고 오류를 반환합니다. 이 경우 fileName 파라미터로 새 이름을 지정하세요.

  • category에 .. 등 상위 폴더로 빠져나가는 경로는 자동으로 제거되어 항상 destRoot 하위에만 저장됩니다.

  • 텍스트/이미지가 아닌 스캔본(이미지 전용 PDF)은 paper_read_text로 텍스트가 거의 추출되지 않을 수 있습니다(OCR 미지원).

Available Tools

3 tools
paper_moveA

분류가 결정된 논문 PDF를 destRoot 아래 category 경로로 이동합니다. category는 'NLP/Sentiment-Analysis'처럼 '/'로 하위 폴더를 표현할 수 있으며, 존재하지 않는 폴더는 자동 생성됩니다. 대상 경로에 동일 이름의 파일이 이미 있으면 덮어쓰지 않고 오류를 반환하니, 필요하면 fileName으로 새 이름을 지정하세요.

ParametersJSON Schema
NameRequiredDescriptionDefault
categoryYesdestRoot 기준 하위 분류 경로. 예: 'NLP', 'CV/Object-Detection' 등 자유롭게 지정
destRootYes분류된 논문을 저장할 루트 폴더의 절대 경로
fileNameNo저장할 파일명을 바꾸고 싶을 때만 지정 (생략 시 원본 파일명 유지)
filePathYes이동할 PDF 파일의 절대 경로 (paper_scan_folder 결과의 path)

TDQS

A4.4/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description bears full responsibility. It discloses key behaviors: auto-creates missing folders, returns error on duplicate files (no overwrite), and supports renaming via fileName. It does not mention permissions or side effects like deletion of original, but 'move' implies that. Overall, sufficient transparency.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single, well-structured paragraph: it starts with the core purpose, then adds details on path syntax, folder creation, overwrite behavior, and the rename option. Every sentence contributes, no redundancy.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool with no output schema and 4 parameters, the description covers parameter usage, edge cases (file conflict, auto-creation), and expected behavior. It might be improved by mentioning what happens to the source file, but 'move' is clear. Overall, adequately complete.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, and description adds meaningful context: category format with example ('NLP/Sentiment-Analysis'), clarification that destRoot is an absolute path, filePath is from paper_scan_folder, and behavior of fileName (optional, to avoid conflicts). This adds value beyond schema definitions.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the action ('move paper PDF'), the resource ('classified paper PDF'), and the destination ('under destRoot and category path'). It distinguishes from siblings by specifying that this handles moving after classification, while paper_read_text and paper_scan_folder serve different purposes.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description specifies when to use the tool ('when paper classification is decided') and provides guidance on handling file conflicts (suggesting fileName to avoid overwrite errors). However, it does not explicitly contrast with sibling tools or state conditions when not to use it.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

paper_read_textA

PDF 파일의 앞부분 페이지에서 텍스트와 메타데이터(Title/Author 등)를 추출합니다. 제목/저자/초록을 확인해 분류 카테고리를 판단하는 데 사용하세요. 분류 로직 자체는 제공하지 않으며, 반환된 텍스트를 바탕으로 호출자가 직접 판단해야 합니다.

ParametersJSON Schema
NameRequiredDescriptionDefault
filePathYes텍스트를 추출할 PDF 파일의 절대 경로
maxCharsNo반환할 텍스트의 최대 문자 수 (기본 6000)
maxPagesNo추출할 앞부분 페이지 수 (기본 3, 보통 제목/초록은 첫 1~2페이지에 있음)

TDQS

A4.2/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations exist, so the description bears full burden. It correctly indicates read-only behavior (extracting from front pages) and that classification logic is not provided. However, it omits error handling, permissions, or file existence details.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two sentences: first defines functionality, second explains use case. No wasted words, clearly front-loaded.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Despite no output schema, description mentions returned text and metadata (Title/Author). It covers the main use case (classification). Could specify return format but overall adequate for a simple extraction tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, baseline 3. Description adds value by stating defaults (maxChars=6000, maxPages=3) and noting that title/abstract are usually in first 1-2 pages, aiding parameter tuning beyond schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Description clearly states the tool extracts text and metadata from front pages of PDFs, and specifies it is used to check title/author/abstract for classification. It distinguishes from sibling tools (move, scan) which have different purposes.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Description explicitly advises using this tool to check title/author/abstract for classification category. It does not provide when-not-to-use or alternatives, but the context is clear given sibling tools are unrelated.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

paper_scan_folderA

지정한 폴더에서 논문 PDF 파일 목록을 스캔합니다. 분류/이동 대상을 찾을 때 가장 먼저 호출하세요. 이미 다른 위치로 이동된 파일은 더 이상 이 폴더에 나타나지 않으므로, 처리 여부를 별도로 추적할 필요가 없습니다.

ParametersJSON Schema
NameRequiredDescriptionDefault
folderYes스캔할 폴더의 절대 경로
recursiveNo하위 폴더까지 재귀적으로 스캔할지 여부 (기본 true)

TDQS

A4.3/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden. It discloses that the tool lists files non-destructively and that files moved elsewhere will no longer appear, implying idempotent behavior without separate tracking. This adds useful context beyond the basic scanning functionality.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is two sentences with no unnecessary words. It front-loads the main action and follows with usage guidance and a behavioral note, achieving maximum efficiency.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's simplicity (2 parameters, no output schema), the description fully covers purpose, usage workflow, and a key behavioral trait (disappearance of moved files). No gaps exist for an AI agent to misinterpret.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% as both parameters have descriptions. The description restates the folder concept ('지정한 폴더') but does not add extra semantic details beyond the schema, so a baseline score of 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool scans a folder for paper PDF files, using specific verbs ('스캔합니다' - scans) and resources ('논문 PDF 파일 목록'). It distinguishes itself from siblings (paper_move, paper_read_text) by being the first step in classification/move workflow.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explicitly instructs to use this tool first when finding classification/move targets ('가장 먼저 호출하세요'). It also explains that moved files disappear, eliminating the need for separate tracking. However, it does not explicitly state when to use alternatives.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. 3 tool updatesv0.1.0
    • First observedpaper_move
    • First observedpaper_read_text
    • First observedpaper_scan_folder
Install Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A4.4/5.0

Scored across 3 tools

Disambiguation5/5

Each tool has a distinct purpose: scanning folders, reading PDFs, and moving files. No overlap in functionality.

Naming Consistency5/5

All tools follow the 'paper_verb_noun' pattern (paper_scan_folder, paper_read_text, paper_move), perfectly consistent.

Tool Count4/5

Three tools is minimal but covers the essential workflow of scanning, reading, and moving papers. Could be expanded, but not under-scoped.

Completeness5/5

The tool set fully covers the paper organization lifecycle: discover papers (scan), extract metadata (read), and categorize (move). No obvious gaps.

Maintenance

ActivityStale
ResponsivenessNo issues

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

Related MCP Servers

  • vectorise-mcp

    jameslovespancakes
    A
    license
    Not graded
    quality
    D
    maintenance
    Local MCP server that indexes folders of documents into a hybrid vector + keyword search index for Claude Desktop, with support for PDFs, Office files, and images via OCR.
    MIT

  • renfield-mcp-filesystem

    ebongard
    A
    license
    Not graded
    quality
    C
    maintenance
    Watches local and SMB folders for settled new files and pushes them into the Renfield knowledge base and Paperless over REST; also provides interactive MCP tools for browsing and on-demand file ingestion.
    MIT

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/iapke486-arch/paper-organizer-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server