Reducto Documentation

Reducto is an agentic document platform for AI teams that turns unstructured documents into data for tools and AI agents. It parses complex layouts, tables, formatting, and bounding-box citations; extracts schema-defined fields; splits multi-document files and long forms into usable units; and edits detected blanks, tables, and checkboxes without predefined templates. Its r-1 parsing model is presented as a single-model alternative to multi-tool document pipelines, while outputs can be reviewed and corrected in real time. Reducto also provides an MCP server and supports document-processing workflows involving parsing, classification, splitting, and extraction.

Reducto Studio is a self-serve document-processing interface for parsing documents, generating and testing extraction schemas, and deploying configured extraction pipelines through an API call or code snippet. Users can upload documents, inspect parsed content, define fields including nested fields and citations, and review extracted JSON records. The workflow supports difficult PDFs such as handwritten or heavily redacted documents and can use OCR, PDF metadata, and figures; the demonstrated evaluation process starts with a small sample and then builds a representative benchmark before scaling.

Reducto MCP Server connects AI agents to Reducto's document-processing APIs through the Model Context Protocol. MCP clients such as Claude Desktop, Claude Code, Codex, Cursor, VS Code, and Windsurf can use it to upload, classify, parse, extract, split, and edit documents, chaining those operations in an agent reasoning loop without custom integration code. It is available as a hosted HTTP server at mcp.reducto.ai/mcp or as a local server run with uvx. The hosted option accepts public URLs, while the local option can read local file paths and public URLs and uses Reducto credentials. The server is described in the accompanying talk as an agent-oriented interface with tools that represent document pipeline work, a snapshot of live pipeline state, confidence scores, and an ask-user mechanism intended to avoid silent guesses.

RE:Dox is an open-source, Apache-2.0-licensed structured-data engine for .NET, developed by CAPCOM as a core component of its REX technology. It parses JSON and other supported formats into a compact token DOM/intermediate representation in which fixed-size tokens can retain source offsets, lengths, links, or inline values; this representation supports reading, editing, serialization, deserialization, and format conversion. The mutable token model provides node-like insert, remove, and replace operations without rebuilding a managed object tree, while lazy decoding can defer materializing strings, numbers, timestamps, and binary values. Its asymmetric I/O design writes known values directly through a DataWriter, while deserialization first builds an addressable token DOM for look-ahead, out-of-order constructor binding, reference resolution, and optional automatic parallel processing of large arrays. The core packages support JSON, JSON5, CBOR, MessagePack, INI, and the binary DOX format, with TOML, XML, HTML, and CSV integrations listed as preview components. A common DataReader/DataWriter and DataConverter model enables conversion between formats such as JSON, CBOR, MessagePack, TOML, XML, and DOX, subject to structural projection where formats have different data models. JSON5 editing can preserve comments and other trivia, and JsonSequence supports asynchronous processing of NDJSON or large top-level JSON arrays without loading the entire input. Compatibility packages are provided for System.Text.Json, Newtonsoft.Json, and DataContractJsonSerializer. The project requires .NET 10 or later, is distributed through CAPCOM.REDox.* NuGet packages, and is under active development; its README notes that APIs, package boundaries, and preview-format support may change before the first stable public release.

doc7 is an open-source document-to-Markdown tool that converts PDFs, Office files, scans, screenshots, charts, formulas, and diagrams into AI-ready Markdown using an OpenAI-compatible multimodal vision model. It renders and understands complete pages rather than relying on character extraction, preserving text, tables, formulas, chart and diagram relationships, image meaning, and visible UI state; the video also describes page-level retry and resume. The model can run locally through LM Studio or Ollama or through a remote endpoint, with no required OCR stack or document-processing service. The same binary provides an interactive CLI, batch processing, model checks, MCP support, a Go SDK, and an asynchronous HTTP service, with installers for macOS, Linux, and Windows.