Actilot — AI Browser Agent

Turn everyday browser tasks into actions with Actilot, an AI agent that works in your browser. Describe your goal in plain language, connect your own AI provider, and follow the task from the extension panel. WHAT YOU CAN DO • Ask the agent to navigate websites, find information, compare options and interact with supported page controls. • Follow task objectives, plans, progress, actions and results in one place. • Answer clarification questions and step in when a website needs your help. • Attach images and text files to give the agent context. • Create and download text files and Excel workbooks from your tasks. • Keep drafts and task history in your browser profile. YOUR PROVIDER, YOUR CHOICE Connect OpenRouter, an OpenAI-compatible endpoint, Ollama, LM Studio or a custom compatible endpoint. You supply the provider and model; provider API charges may apply. Local endpoints require a compatible model with tool support. CONTROL HOW THE AGENT ACTS Ask-before-acting is the default. Guarded Auto allows ordinary reversible actions while requesting confirmation for sensitive or uncertain actions. Autonomous mode executes actions without confirmations, including sensitive actions, within configured task limits. You can pause, stop or resume tasks and manage website access. LOCAL STORAGE, DIRECT AI CONNECTIONS Actilot stores settings, drafts, task history, attachments and generated files in your browser profile. There is no Actilot cloud account or intermediary task-processing backend, and the extension includes no analytics, advertising or tracking SDK. When you run a task, your prompt and relevant website or attachment context are sent directly to the AI provider you configure. That provider's privacy policy and pricing apply. Website access is requested as needed. Optional visual assistance can send page screenshots to a compatible model. Optional voice dictation uses your browser's speech recognition service, which may process audio online. GET STARTED 1. Open Actilot from the extension toolbar or press Alt+Shift+A. 2. Describe a task and connect your provider when prompted. 3. Choose your autonomy level and grant access to the websites needed for the task. 4. Review progress and results in the panel. AI output and website automation can be inaccurate. Some websites, logins and CAPTCHAs require manual intervention. Review results and consequential actions carefully. PDF and Office document attachments are not currently supported.