➔ Omni-Agent is a fully autonomous browser automation and inspection assistant. It can navigate, interact, and execute JavaScript on webpages. It's a powerful tool for anyone who needs to automate browser tasks.

➔ Core Capabilities:

Autonomous Autopilot Mode: Provide the agent with a natural-language prompt (e.g., "Find all broken images and replace them with placeholders") and observe it generate, approve, and execute JavaScript directly on the page.

➔ Deep Context Injection:

With one click, feed the AI your exact browser state. Omni-Agent can instantly read your live DOM Structure, selected Elements, Network Activity, and hooked Console Logs.

Advanced State Management: Full read/write access to Cookies, Local/Session Storage, and IndexedDB. Ask the agent to clear specific session tokens.

➔ Important:

(Data is only sent via user-initiated prompts!)

You are not locked into one ecosystem. Securely store your API keys locally and route tasks to your custom endpoints. Connect your API keys to use at your own limits.

➔ Your code is your code:

Omni-Agent uses a Bring-Your-Own-Key architecture. API keys are encrypted and stored strictly in your local device storage. Code and context are only sent to the LLM provider you explicitly choose, exactly when you click send.

>> Use Omni-Agent from the sidebar menu, F12 tabs, right-click menu, or search bar.