Runs entirely on your machine

Your private
AI intelligence workspace

Upload documents and images. Get structured AI analysis — summaries, risks, entities, timelines. Everything local, everything private.

$ bash start.sh
──────────────────────────────────
Node v22.0.0
Ollama found
Database up to date
Model: phi4-mini — already available
Vision model (llava-phi3) ready
Ready!
 
Operator is running at http://localhost:3000
Everything in one place

A complete workspace for document analysis, investigation, and team intelligence — all scoped to your projects.

Projects

Organise work into named projects. All documents, analysis, and AI chat are scoped to the active project.

Document analysis

AI extracts summaries, metrics, risks, opportunities, and follow-up questions from every upload with domain-specific framing.

Image uploads

Upload photos and screenshots — AI describes the scene, extracts text (OCR), and reads EXIF metadata including GPS coordinates.

Dispatch

AI chat with mode-specific personas. Full access to your document context, web search, and persistent memory across conversations.

Inspector & story map

Click any entity or map pin to open a panel showing source documents and connected entities without leaving the page.

Pulse

Feed aggregator for RSS, Reddit, YouTube, Bluesky, and Mastodon — with keyword monitoring and automatic refresh.

File management

Browse and manage files saved by Operator, organised by project and area. Analyse documents directly from the folder view without re-uploading.

Metrics board

KPI board aggregating every metric extracted across all documents, grouped by area with source attribution.

Remote submissions

Share a link so others can submit files from any device. Files land directly in your library — no app required.

Local-first, always

Operator runs entirely on your machine using Ollama. No data leaves your device unless you explicitly choose a cloud AI provider. API keys are encrypted with AES-256-GCM and stored locally. No analytics, no telemetry, no tracking.

Adapts to your domain

Switch modes in Settings to change the interface, AI framing, terminology, and feature set to match your work.

Executive

Business reporting, team oversight, KPI tracking across areas.

Journalism

Field notes, sources, entities, timelines, redactions, and story building.

Team Lead

Sprint updates, blockers, retrospectives, and team health tracking.

Market Research

Interview transcripts, survey data, theme discovery, and pattern analysis.

Legal

Case files, evidence chains, matter management, entity and timeline extraction.

Human Resources

Workforce analytics, engagement tracking, HR reports, and people insights.

Up and running in minutes

One script handles everything — Node.js, Ollama, the database, and AI models.

1

Download

Download the ZIP and unzip it somewhere permanent — your home folder, Documents, wherever makes sense. Or clone with git if you prefer.

2

Run the start script

One command. On first run it installs all dependencies and pulls the default AI models. Every subsequent run starts in seconds.

3

Complete onboarding

Choose your appearance, pick an app mode, and enter your name and role so the AI can tailor its responses.

4

Create your first project

Name it, set an area and start date, then upload your first documents. All analysis is scoped to the active project.

# Option A — download ZIP (no git required)
curl -L https://github.com/jcktp/operator/releases/latest/download/operator.zip -o operator.zip
unzip operator.zip && cd operator
 
# Option B — clone with git
git clone https://github.com/jcktp/operator.git && cd operator
 
# Run (handles everything on first launch)
bash start.sh
 
# Opens automatically at http://localhost:3000
# Requirements: macOS or Linux · ~5 GB disk · 8 GB RAM

Ollama is the default and runs fully offline. Switch to any cloud provider in Settings — keys are encrypted locally and never shared.

Ollama (default) Anthropic Claude OpenAI GPT Google Gemini Groq xAI Grok Perplexity Mistral