"How to read document contents" matching MCP servers:

Convert PDFs to structured JSON. Extract invoices, bank statements, contracts, and more. Pay per call via x402 USDC.

Provides read-only, authenticated access to prepared Markdown, images, document structure, and original-file references published as immutable versioned bundles alongside Paperless-ngx source documents. It can also explicitly forward selected prepared Markdown into WeKnora's existing ingestion interface without subscribing to future documents.

Enables conversion between PDF/Office documents and structured JSON, supporting extraction and generation of document content.

Enables agents to reconcile piping material take-offs against Plant 3D SCSA exports by previewing diffs and applying quantity corrections, insertions, and gated removals with dry-run and explicit confirmation flags.

Enables checking lead rows, CSV files, or Apify datasets for duplicates, email syntax and risk signals, domain issues, missing fields, and stale dates, returning a per-row import, review, or reject verdict so only clean records reach a CRM or sequencer.

Enables document processing with Suparse API, extracting structured data from documents into JSON, CSV, XLSX, or Google Sheets, and supports processing single files or folders.

Offline-first MCP server for TwitterAPI.io docs (endpoints, pages, blogs) with fast fuzzy search.

MCP server that provides AI assistants with tools to understand, validate, generate, and transform CERIF 1.6 research information, including entity browsing, XML validation, document building, and migration from CERIF 1.5.

Enables AI assistants to search multiple tech job boards at once, summarize what the market is asking for, and rank postings by how well they match a user's own skills.

A Model Context Protocol (MCP) server that enables MCP clients like Claude Desktop to interact with protocols.io, a popular platform for sharing scientific protocols and methods.

The Excel to JSON MCP provides a standardized interface for converting Excel and CSV data into JSON format using the Model Context Protocol.

Enables an AI assistant to query and browse multiple database engines read-only, with the ability to switch connections and engines dynamically during a session.

Enables document processing through Upstage AI services including parsing various document formats, extracting structured information with custom schemas, auto-generating extraction schemas, and classifying documents into categories.

Gives Claude read-only tools to list leads, inspect normalized business context, preview the deterministic routing decision (including a country inferred from untrusted notes, always flagged for review), inspect a case's ledger state and audit trail, and explain the ordered policy rules. All outputs are read-only, so the assistant can explain why a lead went where it did while every assignment still requires human approval.

Enables agents to extract multi-page financial tables from complex PDFs, rasterize vector pages into crisp 300 DPI images for vision LLMs, flatten interactive AcroForm/XFA forms, and dewarp skewed OCR bounding boxes from photographed or scanned paper documents. Tool invocations are gated by USDC micropayments settled on Base L2 via the x402 protocol.

Read-only-by-default MCP server for Datanika, an open-source dlt + dbt-core data pipeline platform: browse connections, introspect and preview source tables, compile and run dbt transformations, and monitor pipeline runs. 25 tools (17 read-only, plus 8 write tools gated behind --allow-write); source lives in the datanika-mcp/ subdirectory of the repo.

The JSON to Excel MCP provides a standardized interface for converting (1)JSON data (2)URL pointing to JSON files into CSV format

Enables users to clean and de-duplicate sales lead lists locally without API keys or network calls. It normalizes names, emails, phone numbers, and website domains, drops invalid emails, removes duplicate leads by email, and validates email formatting.

Converts JSON data to TOON (Token-Oriented Object Notation) format and back, reducing token usage by 30-60% for more efficient LLM applications.

Enables AI assistants to extract and structure content from documents (PDFs, images, Office files) using Upstage AI's document digitization and information extraction APIs.

Enables read-only access to FileMaker databases through the Data API, allowing users to retrieve records, analyze metadata, search across layouts, and infer relationships while maintaining data security.

A server that connects to PostgreSQL databases and provides tools for safely exploring schemas, running read-only SQL queries, and performing data analysis with pre-built templates.

Enables retrieval and analysis of graph data from the Pythagraph RED API. Provides formatted tables, statistics, node/edge distributions, and comprehensive summaries for graph visualization and insights.

Local-first Taiwan market evidence workbench for AI agents with persistent Watchlists, governed MCP, auditable evidence, and explicit execute-once retrieval. GitHub Repository URL

MCP server for Lakehouse42, enabling code-first tool discovery, hybrid search, document management, and Iceberg time-travel queries with optimized responses.

Enables agents to read and stage geospatial datasets via Geo Hub, with read-only and controlled staging tools.

Enables AI agents to safely ingest, inspect, edit, and export manufacturing documents (Excel, PDF, Word, Markdown) with controlled patch workflows and MES entity extraction.

Enables document conversion to Markdown or OKF format, and provides 14 okf_* tools for OKF bundle operations via MCP.

Enables agents to transform tabular Markdown content into structured, type-inferred records with normalized formatting through an MCP-compatible interface.

MCP server that exposes Parseur document parsing and mailbox management as tools for AI assistants, allowing natural language control of parseur.com.