Rosply
Rosply is an AI agent that automates desktop tasks via natural language instructions.
FreeProductivity
Visit website Free to browse · updated 2026-06-23
What is Rosply?
Rosply operates by capturing screenshots at each step to analyze the current display and determine the next actions needed. Users provide tasks in everyday language, after which the agent manages sequences of inputs across applications. Setup involves running a script that configures the environment and required models. Key capabilities include persistent memory for carrying information between steps, an emergency stop function, and compatibility with various AI providers. It also features local wake-word detection for hands-free operation and direct file writing within code editors. Common applications range from web form handling and data extraction to file organization and spreadsheet automation. The agent works on both simple and multi-step workflows while maintaining limits on action counts for safety.
Key features
Vision-powered automation using screenshots every step
Full desktop control including click, type, scroll, and hotkeys
Voice control with local Whisper wake word "Hey Rosply"
VS Code code generation via built-in extension
Persistent memory across task steps and runs
Emergency stop with Ctrl+H and 200-action cap
MCP integration with Claude Code and support for OpenRouter, Ollama, GPT-4o
AI models Rosply uses
Qwen VL
Default model via OpenRouter
DeepSeek V3
Via OpenRouter for VS Code code generation
claude-sonnet-4-6
Claude Sonnet 4.6 via Claude Code MCP
Fable 5
Via Claude Code
Gemini
Any vision API
GPT-4o
Any vision API
What you can use Rosply for
Automate Invoice Downloads
Rosply navigates billing portals, identifies invoices on screen via vision, scrolls as needed, and downloads files without scripts or DOM access.
Fill Job Application Forms
Type a plain-English instruction and Rosply clicks fields, enters data, and submits web forms across dynamic interfaces using full desktop control.
Scaffold Projects in VS Code
Rosply generates code files directly in the editor via its built-in extension, routing prompts to models like DeepSeek V3 through OpenRouter.
How to use Rosply
- 1Run the setup script to install Python environment, Whisper model, and VS Code extension
- 2Create a free OpenRouter account and paste the API key into the .env file
- 3Launch Rosply and type your task in plain English
- 4Press Enter and watch the agent capture the screen and execute actions
- 5Press Ctrl+H at any time for an instant emergency stop
Rosply pricing
Pricing model: Free. Plan details are indicative — check the site for current prices.
Free
Free
- PC Automation Agent
- Vision-Powered control
- Full Desktop Control
- VS Code integration
- Voice Control
- Persistent Memory
- Emergency Stop
- OpenRouter / local models support
Editor's verdict
Pros
- +No scripts or brittle selectors required
- +Works on any app including those without APIs
- +Supports local models and multiple vision providers
Cons
- –Requires Python 3.11+
- –macOS support is beta only
- –Needs external API key (e.g. OpenRouter) for default operation
Our take: Rosply is a solid productivity choice. It's valued for no scripts or brittle selectors required and works on any app including those without apis. The main trade-off is requires python 3.11+. A good pick if you want capable AI without a high upfront cost.
Frequently asked questions
Rosply runs on Windows 10/11 and Linux fully, with macOS available in beta; Python 3.11 or higher is required.
Summary
Rosply is a solid productivity choice. It's valued for no scripts or brittle selectors required and works on any app including those without apis. The main trade-off is requires python 3.11+. A good pick if you want capable AI without a high upfront cost.
Did you find this helpful?
User reviews
Verified reviews from the community shape this tool's rating.
Loading reviews…
Rosply alternatives
Similar productivity tools worth comparing.
