Features

Every feature. Every detail.

WindOp packs 35+ tools into a single AI assistant. Here's everything it can do.

AI & Models

300+ Models via OpenRouter

Access GPT-4, Claude, Gemini, DeepSeek, Llama, and hundreds more through a single API.

Smart Task Categorization

Auto-selects the optimal model for each task type — coding, writing, analysis, creative work.

Customizable Preferences

Set favorite models per task type. Quick switch with Ctrl+M. Reasoning token display for o1/o3 models.

Desktop Automation

Screenshot Analysis

AI vision captures and understands your screen. Find elements, read text, analyze layouts.

Mouse & Keyboard Control

Click, scroll, drag, type, send shortcuts. Full input automation through natural language.

Application Management

Launch, switch, and manage applications. Auto-detect installed software.

File Operations

Read, Write, Edit, Delete

Full filesystem access. Smart find-and-replace with fuzzy matching. Multi-edit in one call.

Semantic Diffs

Aider-style SEARCH/REPLACE blocks. Automatic backups before every edit. Preview with syntax highlighting.

Search Everywhere

Recursive file search. Content search across directories. Glob patterns and regex support.

Shell & Terminal

Multi-Shell Support

PowerShell, CMD, WSL — auto-detects the right shell. Smart command correction (Unix → Windows).

Real-Time Streaming

Watch output as it happens. Background process management. Interactive command blocking for safety.

Safe Execution

Confirmation dialogs for destructive commands. Per-tool permission declarations.

Web & Research

Multi-Provider Web Search

Search the web directly from chat. Extract and summarize page content. Open URLs in your browser.

Content Extraction

Jina Reader integration for clean article extraction. Readability processing strips clutter.

Research Workflows

Combine search, extraction, and summarization in multi-step research tasks.

Image Generation

Text-to-Image

Generate images with Flux, Seedream, Riverflow. Style presets: photorealistic, anime, oil painting, more.

Image Editing

Upload images and modify them with AI. Add text, change styles, transform compositions.

High Resolution

1K, 2K, and 4K output. Multiple aspect ratios: 1:1, 16:9, 4:3, 3:4, and more.

Memory & Context

Persistent MEMORY.md

Categorized memory: preference, knowledge, pattern, insight. Loaded every session automatically.

Daily Logs

Timestamped entries for events, decisions, and learnings. Full session history.

Smart Context Management

Semantic relevance scoring. Only loads what matters for the current task.

Multi-Agent Orchestration

Group Chats

Create specialist agents with custom roles. Planner, reviewer, security analyst — all in one conversation.

Teams

Supervisor-coordinated teams. Parallel or sequential execution. Discussion mode for refinement.

Complex Workflows

Explore → Code → Review pipelines. Parallel research. Multi-perspective debate.

Security & Privacy

Local-First

Everything runs on your machine. No cloud processing. Your conversations never leave your PC.

Zero Telemetry

No analytics. No tracking. No phone-home. Period.

Safe by Design

Confirmation dialogs for destructive actions. Automatic file backups. API keys in secure storage.

UI & Experience

Modern Dark Theme

Refined dark aesthetic with smooth animations. Clean, minimal interface built with Framer Motion.

Power User Features

Dense mode. Customizable keybindings. Token usage visualization. Agent status indicators.

Responsive Design

Works at any window size. Clean, modern, minimal interface that stays out of your way.