Skip to main content
Glama
rexfelix

PDF Reader MCP Server

by rexfelix
OverviewSchemaRelated ServersScoreDiscussions
Python
Local

PDF Reader MCP Server

1. 개요

이 문서는 PDF Reader MCP Server를 설치하고 실행하며, 실제 사용하는 방법을 안내합니다. 이 서버는 PDF 문서를 읽고, 텍스트(Markdown), 이미지, 표를 추출하여 AI에게 제공하는 기능을 수행합니다.

Related MCP server: PDF Reader MCP Server

2. 설치 및 실행 (Installation & Execution)

이 프로젝트는 uv 패키지 매니저를 사용하여 의존성을 관리합니다.

2.1. 필수 요구사항

  • uv (Python Package Manager)

  • Tesseract-OCR (OCR 기능 사용 시 필요)

    • macOS: brew install tesseract tesseract-lang

    • Linux (Ubuntu): sudo apt-get install tesseract-ocr tesseract-ocr-kor

    • Windows: UB-Mannheim/tesseract 설치 관리자 다운로드 후 설치 시 'Korean' 언어 팩 선택

2.2. 환경 설정 (uv)

프로젝트 루트에서 다음 명령어로 가상환경을 생성하고 의존성을 설치합니다.

uv sync

2.3. 환경 점검

설치된 환경(uv 가상환경)이 올바른지 확인합니다.

uv run src/utils/env_check.py

2.4. 서버 실행 테스트

터미널에서 직접 실행하여 에러가 없는지 확인합니다.

uv run src/server/main.py

3. MCP 서버 등록 (Claude Desktop / Antigravity)

클라이언트 설정 파일(claude_desktop_config.json 등)의 mcpServers 섹션에 아래와 같이 추가하십시오.

{
  "mcpServers": {
    "pdf-reader": {
      "command": "uv",
      "args": [
        "--directory",
        "/YOUR/ABSOLUTE/PATH/TO/pdf_reader_mcp_server",
        "run",
        "src/server/main.py"
      ]
    }
  }
}

참고: args의 /YOUR/ABSOLUTE/PATH/TO/... 부분을 실제 프로젝트가 위치한 절대 경로로 변경해야 합니다.

4. 기능 사용 가이드 (Usage)

4.1. MCP Tools

AI 에이전트는 다음 도구들을 사용하여 PDF와 상호작용합니다.

도구 이름

설명

인자 예시

read_pdf

PDF 내용을 읽어 Markdown으로 반환

source="/path/to/doc.pdf", page_range="1-3"

get_pdf_metadata

제목, 페이지 수 등 메타데이터 확인

source="https://example.com/paper.pdf"

4.2. MCP Resources

URI를 통해 직접 접근할 수도 있습니다.

  • URI Pattern: pdf://{absolute_path_to_file}

  • 사용법: AI에게 "이 파일 읽어줘: pdf:///Users/me/docs/manual.pdf" 라고 요청하면, 내부적으로 read_resource를 호출하여 처리합니다.

4.3. Prompts

사전 정의된 프롬프트를 사용할 수 있습니다.

  • summarize_pdf: 지정된 PDF를 읽고 요약해줍니다.

5. 테스트 결과 (Verification)

자동화된 단위 테스트를 통해 핵심 모듈(Loader, Extractor, Parser)의 동작을 검증했습니다.

  • 로컬 파일 및 URL 로딩 테스트 완료

  • 텍스트 및 이미지 추출 테스트 완료

  • 통합 파싱 로직 테스트 완료

uv run python -m unittest discover tests
# Result: OK

Available Tools

2 tools
get_pdf_metadataB

Quickly retrieve metadata from a PDF without reading the full content.

ParametersJSON Schema
NameRequiredDescriptionDefault
sourceYes

Output Schema

ParametersJSON Schema
NameRequiredDescription
resultYes

TDQS

B3.3/5.0
Behavior2/5

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

With no annotations provided, the description carries the full burden of behavioral disclosure. It mentions 'quickly retrieve' which hints at performance, but doesn't cover critical aspects like error handling, permissions, rate limits, or what specific metadata is returned. This leaves significant gaps for a tool with an output schema but no annotation support.

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 sentence that efficiently conveys the core purpose and key differentiator. It's front-loaded with the main action and avoids any unnecessary words, making it highly concise and effective.

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

Completeness3/5

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

Given the tool has an output schema (which handles return values), no annotations, and low complexity, the description is minimally complete. It covers the basic purpose and hints at usage, but lacks details on behavior and parameters. This makes it adequate for a simple tool but with clear gaps in guidance and transparency.

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?

The input schema has 1 parameter with 0% description coverage, so the description must compensate. It doesn't add any details about the 'source' parameter beyond what the schema provides (e.g., format, examples, constraints). Since there's only one parameter and the description doesn't enhance its meaning, the baseline is 3, as it's adequate but doesn't add value beyond the schema.

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

Purpose4/5

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

The description clearly states the tool's purpose with a specific verb ('retrieve') and resource ('metadata from a PDF'), and distinguishes it from the sibling tool 'read_pdf' by specifying 'without reading the full content'. However, it doesn't explicitly name the sibling tool for comparison, keeping it at a 4 rather than a 5.

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

Usage Guidelines3/5

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

The description implies when to use this tool by contrasting it with reading full content, suggesting it's for quick metadata extraction. However, it doesn't provide explicit guidance on when to choose this over 'read_pdf' or any alternatives, nor does it mention prerequisites or exclusions, so it's only implied usage.

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

read_pdfA
Read content from a PDF file (local path or URL).
Returns a unified Markdwon string containing text, tables, and image references.

Args:
    source: Local file path or URL to the PDF.
    page_range: Format "1-5" or "10". If not provided, reads all pages.
    extract_images: If True, extracts images to temp dir and links them.
ParametersJSON Schema
NameRequiredDescriptionDefault
sourceYes
page_rangeNo
extract_imagesNo
force_ocrNo

Output Schema

ParametersJSON Schema
NameRequiredDescription
resultYes

TDQS

A4.2/5.0
Behavior4/5

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

With no annotations provided, the description carries the full burden of behavioral disclosure. It effectively describes key behaviors: it reads PDF content, returns a unified Markdown string, handles local/URL sources, supports page ranges, and can extract images to temp dir with links. However, it doesn't mention potential limitations like file size restrictions, supported URL types, or error handling.

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 appropriately sized and front-loaded: the first sentence states the core purpose, followed by return value, then parameter details in a clear 'Args:' section. Every sentence adds value without redundancy, making it easy to scan and understand.

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?

Given the tool has 4 parameters with 0% schema coverage and an output schema exists, the description does well by explaining most parameters and the return format. However, it misses documenting the 'force_ocr' parameter entirely, and with no annotations, it could benefit from more behavioral context like performance characteristics or error conditions.

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?

The description adds significant meaning beyond the input schema, which has 0% description coverage. It explains that 'source' can be a local path or URL, 'page_range' uses format '1-5' or '10' and defaults to all pages, and 'extract_images' extracts images to temp dir and links them. However, it completely misses the 'force_ocr' parameter, leaving it undocumented.

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 specific action ('Read content from a PDF file') and resource ('PDF file'), distinguishing it from the sibling tool 'get_pdf_metadata' which presumably retrieves metadata rather than content. The description explicitly mentions what it reads (text, tables, image references) and returns (unified Markdown string).

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

Usage Guidelines3/5

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

The description implies usage by stating it reads PDF content and returns Markdown, but doesn't explicitly say when to use this tool versus alternatives like 'get_pdf_metadata'. It mentions the tool can handle both local paths and URLs, which provides some context, but lacks explicit guidance on when to choose this over other PDF-related tools or exclusions.

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. 2 tool updates
    • First observedget_pdf_metadata
    • First observedread_pdf
Install Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A3.6/5.0
Disambiguation5/5

The two tools have clearly distinct purposes: get_pdf_metadata retrieves metadata without reading content, while read_pdf extracts the actual text, tables, and images. There is no overlap or ambiguity between these functions.

Naming Consistency5/5

Both tools follow a consistent verb_noun pattern (get_pdf_metadata and read_pdf), using clear action verbs and the same domain prefix. The naming is predictable and readable throughout.

Tool Count2/5

With only 2 tools, the server feels thin for a PDF reader domain. While the tools cover basic reading and metadata, there are likely missing operations like searching within PDFs, extracting specific elements, or handling annotations, which limits functionality.

Completeness2/5

The tool surface is significantly incomplete for a PDF reader. It lacks essential operations such as search, annotation handling, page manipulation, or conversion to other formats. Agents will face dead ends when trying to perform common PDF tasks beyond simple reading.

Maintenance

ActivityInactive
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

Appeared in Searches

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/rexfelix/readPDF_mcp_server'

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