This server provides AI-powered PDF analysis via multiple LLM providers, enabling you to ask questions, summarize, extract tables, compare sections, and work with PDFs through natural conversation.
Analyze PDF documents from local file paths, remote URLs, or cached file URIs (Google provider)
Ask multiple questions in a single request
Summarize key points and extract specific data, including tables
Compare sections within a document or across documents
Reuse cached PDF content and handle chunked/large PDFs using cached file URIs (Google provider)
Choose from multiple providers: Google Gemini, Anthropic Claude, OpenAI GPT, Google Vertex AI, Anthropic on Vertex AI
Integrate with AI agents/tools like Claude Code and OpenAI Codex
Run interactively or as a hosted HTTP server with endpoints for analysis (/analyze), MCP (/mcp), and health checks (/health)
Native standalone binaries for macOS, Linux, and Windows with auto-updates and no dependencies
Allows AI agents to analyze PDF documents using Google Gemini's models.
Allows AI agents to analyze PDF documents using OpenAI's GPT models.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@pdf-analyzer Summarize the key points in report.pdf"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
The PDF Analyzer MCP Server gives AI agents the ability to read and analyze PDF documents, enabling document Q&A through natural conversations.
Supports multiple LLM providers: Google Gemini, Anthropic Claude, and OpenAI on their direct APIs, plus Google Vertex AI and Anthropic on Vertex AI for service-account auth. Choose your preferred provider and model during setup.
macOS / Linux:
curl -fsSL https://raw.githubusercontent.com/IntelligentElectron/pdf-analyzer/main/install.sh | bashWindows (PowerShell):
irm https://raw.githubusercontent.com/IntelligentElectron/pdf-analyzer/main/install.ps1 | iexWhy use the native installer:
No dependencies — standalone binary, no Node.js required
Auto-updates — checks for updates on startup
Signed binaries — macOS binaries are notarized by Apple
Platform | Install Directory |
macOS |
|
Linux |
|
Windows |
|
The server checks for updates on startup. To update manually:
pdf-analyzer --updateRelated MCP server: PDF RAG MCP Server
For developers who prefer npm:
npm install -g @intelligentelectron/pdf-analyzerOr use with npx (no installation required):
npx @intelligentelectron/pdf-analyzer --helpRequires Node.js 20+.
To update:
npm update -g @intelligentelectron/pdf-analyzerAfter installing, run the interactive setup to choose your provider, model, and enter your API key:
pdf-analyzer --setupYou'll be prompted to choose from:
Provider | Fast Model | Flagship Model | Get API Key |
Google Gemini | Gemini 3 Flash | Gemini 3.1 Pro | |
Anthropic Claude | Claude Sonnet 4.6 | Claude Opus 4.7 | |
OpenAI GPT | GPT-5.4 Mini | GPT-5.4 |
Claude Opus 4.6 is offered alongside 4.7 as the previous flagship. The Vertex AI providers offer the same Gemini and Claude models, and authenticate with a service account JSON key file instead of an API key.
You can re-run --setup at any time to switch providers or models.
After setup, connect the MCP to your AI agent of choice.
Install Claude Code, then run:
claude mcp add --scope user pdf-analyzer -- pdf-analyzerInstall OpenAI Codex, then run:
codex mcp add pdf-analyzer -- pdf-analyzerOnce connected, ask your AI assistant to analyze any PDF:
"Analyze /path/to/document.pdf and summarize the key points"
"What tables are in this PDF? Extract the data from table 2"
"Compare the findings in sections 3 and 5 of this report"
The server accepts:
Local file paths: /Users/name/docs/report.pdf
URLs: https://example.com/document.pdf
Platform | Binary |
macOS (Universal) |
|
Linux (x64) |
|
Linux (ARM64) |
|
Windows (x64) |
|
Setting PORT starts the server over Streamable HTTP instead of stdio, serving MCP at /mcp, a direct POST /analyze REST endpoint, and GET /health.
See deploy/README.md for deploying it to Cloud Run, including the provider and auth matrix, the IAM roles each provider needs, and how to reach the private service.
See docs/architecture.md for how the server is put together.
See CONTRIBUTING.md for development guidelines.
Created by Valentino Zegna
This project is hosted on GitHub under the IntelligentElectron organization.
Apache License 2.0 - see LICENSE
Analyze a PDF document using AI. Provide an absolute file path, URL, cached file URI (from a previous response, Google only), or array of cached file URIs (from a previous chunked response, Google only) and a list of questions to ask about the PDF content. With the Google provider, returns a cached_uris array that can be reused for subsequent queries on the same document.
| Name | Required | Description | Default |
|---|---|---|---|
| queries | Yes | Array of questions to ask about the PDF | |
| pdf_source | Yes | PDF source: absolute local file path, URL, cached file URI from a previous response (Google only), or array of cached file URIs from a previous chunked response (Google only) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It mentions the caching behavior for Google but fails to describe the primary output (analysis results). The agent does not know what the tool returns beyond an optional cached_uris array.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two concise sentences, each serving a purpose. The description is front-loaded with the main action, followed by input specifics and a note on output. No wasted words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the simplicity (2 params, no output schema, no annotations), the description covers input types and caching but omits the primary output format, error handling, or any limitations. It is partially complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, but the description adds value by clarifying provider-specific details (cached URIs for Google only) and contextualizing the pdf_source options beyond the schema's description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states 'Analyze a PDF document using AI' with specific verb and resource, and distinguishes input types. It is precise and leaves no ambiguity about the tool's function.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
While no sibling tools exist, the description provides clear context on acceptable input formats (file path, URL, cached URI) and hints at provider-specific usage (Google caching). It lacks explicit when-not-to-use or alternatives, but the context is sufficient.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
v1.2.4analyze_pdfWith only one tool, there is no possibility of ambiguity. The tool 'analyze_pdf' stands alone with a clear purpose.
The single tool 'analyze_pdf' follows a clear verb_noun pattern, which is standard and consistent.
A single tool for a PDF analyzer is too few. While the tool itself is non-trivial (AI analysis), the server lacks any additional operations like extraction, merging, or conversion, making it feel incomplete for typical PDF tasks.
The server only offers AI-based analysis of PDFs. There are obvious gaps such as missing CRUD operations for PDF content, text extraction, or file manipulation, which limits its usefulness.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Generate and read PDFs for AI agents: a generate_pdf and a read_pdf tool, priced per document.
Parse, extract, split, and ask over digital PDFs (text layer, no OCR) from Cursor and Claude.
Ingest, manage, and retrieve documents for RAG-powered AI applications
Generate PDFs from templates via AI chat. Works with Claude, ChatGPT, Cursor, and any MCP client.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/IntelligentElectron/pdf-analyzer'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
