MultiTool Office Next
Windows office workbench with search, OCR and batch renaming, where Jev optionally helps pick extracted contract fields.
Overview
MultiTool Office Next is a free, open-source Windows workbench built with Tauri, React and Rust. It combines multi-column file browsing and local search, file previews, an AI office assistant using OpenAI-compatible models, OCR-based field extraction with batch renaming, translation and a to-do manager. The README is in Chinese.
Jev appears in the OCR pipeline: after RapidOCR reads PDF or image documents and rules extract candidate fields such as company name, contract number, date or product, Jev can optionally help judge which candidate is correct. Before candidates are sent, amounts and long numbers are filtered out. Results show source lines and decision records so users can correct them before renaming, which is previewed and undoable. Keys are stored in Windows Credential Manager. It suits office workers who process batches of contracts or receipts.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Choice | Which OCR candidate is the correct value for this contract field?(inferred) | — |
How it works
- Run local OCR on selected PDFs or images.
- Extract candidate field values with tags, regex and layout rules.
- Optionally ask Jev to pick among candidates after sensitive numbers are filtered.
- Review decisions, correct, then preview and apply batch renaming.
Strengths
- Local OCR and indexing by default.
- Sensitive numbers filtered before sending candidates to Jev.
- Undoable batch renaming with decision records.
Caveats
- Jev use requires your own TypeSafe key and is optional.
- Windows only; documentation in Chinese.
- User-submitted text and images are outside the redaction scope.
Getting started
Run the MultiTool Office Next x64 setup installer from GitHub Releases, add a working folder, and configure Jev keys under the network and scoring settings.At a glance
- Type
- Project
- Status
- beta
- Language
- Rust
- Runtime
- Tauri desktop app
- Has code
- yes
- License
- Apache-2.0
Use cases and integrations
How curated lists describe it
Windows file manager and office workbench with local search, OCR, translation and to-dos; Jev helps pick extracted contract fields, with amounts and long numbers filtered out first. Optional, in Chinese.
MrJev/awesome-jev
Frequently asked questions
What does Jev decide here?
Which extracted candidate is the right value for a contract field during OCR processing.
Is document content sent to Jev?
Only candidate information, with amounts and long numbers filtered out first.
Is Jev required?
No, it is an optional assist that you enable with your own key.
Related projects
Docker service that classifies the Paperless-ngx inbox with Jev (or local Clef via Ollama), setting confident fields and queueing the rest.
ProjectbetaExperimental Markdown-to-Sanity mapper where Jev picks which source block fills each field and code copies it, generating no text.
ProjectprototypeTypeScriptPrivacy-first personal knowledge management system, listed as integrating Jev for document classification and tag routing.
PluginproductionTypeScriptClassifies tax PDF pages into 261 IRS forms and 7 page kinds with one Jev request per page and a confidence gate.
ProjectproductionTypeScriptRust tool that asks Jev which PDF pages genuinely need OCR, extracting text pages locally and sending only the rest to OCR.
ProjectbetaRustGo terminal app that files documents using five Jev questions: category, sensitivity, urgency, whether action is needed and action state.
ProjectbetaGo