Unstract
Unstract is a self-hosted tool that pulls structured data out of PDFs and scanned documents using plain-language prompts for finance and compliance teams.
What it is good for
- Extracting fields from invoices and scanned PDFs into a database
- Turning mixed document types into clean JSON without writing templates
- Processing folders of files on a schedule and loading results to a warehouse
- Connecting document extraction to an AI assistant or automation workflow
At a glance
- Who it is for
- Setup
- Guided command-line setup
Level 4 of 5. You copy and paste commands into a terminal by following the instructions. Comfortable territory for vibe coders.
- Cost
- Free tier plus paid plans
- Type
- Agent app
- Tasks
- Works with
- Claude, n8n, OpenAI, Anthropic Claude, Snowflake, Google BigQuery
Momentum
- Rank this week
- Not ranked yet
- Best rank
- None yet
- GitHub stars
- 7,397
GitHub numbers as of Nov 2, 2026. Updated weekly.
Official links
Setup steps are on the official pages. We do not repeat them here, since they change.
Questions
- Is Unstract free?
- It has a free plan and paid plans, according to its own description.
- Do I need to know how to code to use Unstract?
- You do not need to write code, but setup means copying and pasting commands into a terminal by following its instructions.
- What does Unstract work with?
- According to its own documentation, it works with Claude, n8n, OpenAI, Anthropic Claude, Snowflake, and Google BigQuery.
- What is Unstract good for?
- Extracting fields from invoices and scanned PDFs into a database, turning mixed document types into clean JSON without writing templates, processing folders of files on a schedule and loading results to a warehouse, and connecting document extraction to an AI assistant or automation workflow.
Similar agents
- CowAgent
CowAgent is an open-source personal AI assistant that plans tasks, runs tools and skills, and keeps memory for people who want a 24/7 agent on their own computer or server.
- OfficeCLI
OfficeCLI is a command-line tool that lets AI agents read, edit, and create Word, Excel, and PowerPoint files without Office installed.
- gws
gws is a command-line tool that lets you and AI agents manage Google Workspace (Drive, Gmail, Calendar, Sheets, Docs) from one place.
- cli
CLI is a command-line tool that lets people and AI agents work with Lark/Feishu documents, chats, calendars, and mail for business teams.
- DeepBot
DeepBot is a desktop AI assistant that handles enterprise office work such as documents, data, and monitoring for business teams.
- Synaplan Desktop
Synaplan Desktop is a desktop app that pairs your computer with a Synaplan workspace and runs installed Agent Skills locally for people who use Synaplan.