Send tasks from any page
Press ⌘E on any page to send a plain-language request to Claude Code or Codex. Rill uses the page context and routes the request to a relevant local project, asking when it is unsure.
Rill is a free beta Mac browser that lets users send tasks from any web page to Claude Code or Codex agents while continuing to browse. It also organizes tabs and browsing history around ongoing projects, with basic browsing available without an agent.
Rill is a native Mac browser that places Claude Code and Codex agents alongside ordinary web browsing. From any page, you can send a task in plain language, ask about the page, or continue reading while the agent works in the background. Rill can use the Claude Code or Codex plan already associated with your account, so the site says there is no separate Rill subscription or API key requirement.
The browser also treats browsing as an ongoing record of work. It groups tabs, keeps day-based history, finds remembered pages from descriptions, and maps the projects your agents have touched. Rill is free and in beta; AI features require a supported, installed, and signed-in version of Claude Code or Codex, while basic browsing works without an agent.
Press ⌘E on any page to send a plain-language request to Claude Code or Codex. Rill uses the page context and routes the request to a relevant local project, asking when it is unsure.
Run your own Claude Code or Codex CLI in the background while you continue browsing. Tasks can show tool steps and diffs, and work in git repositories can be undone.
Agents can open their own background tabs with separate website data. In project work, they read the web without your logins; you can take over a tab when needed.
Rill groups tabs by what you were doing and can compare up to six tabs in a table that quotes each page. It also works as a browser when no agent is connected.
The browser records activity as day-based episodes, helps find a previously visited page from a description, and displays projects on a map with AI-named areas.
Passwords are stored in the Mac Keychain, while existing sign-ins can be brought over from Chrome, Arc, or Safari. History and passwords remain on the Mac, and tab sorting and day summaries can be disabled.
While reading documentation, an issue, or a reference page, send a request with ⌘E instead of switching to a terminal. Rill places the task with the relevant project and lets you continue reading while the agent works.
Ask an agent to inspect several sources, or select up to six tabs and generate a comparison table that quotes each page. This keeps source review inside the browser rather than splitting it across windows.
Use an agent’s background tabs to browse websites as part of project work without exposing your normal sign-ins. Take over the tab when you need to review or continue the interaction yourself.
Use the browser’s day episodes, description-based search, and project map to reconstruct what you visited and which projects your agents touched. This suits work that unfolds across several sessions.
Use Rill for ordinary browsing when no agent is connected, then ask about a page or start an agent task when a page raises a coding or research question.
Rill is free. Its AI features run through the Claude Code or Codex plan already associated with your account; the site does not provide further pricing or usage-plan details.
For AI features, you need Claude Code 2.1.261 or later, or Codex CLI 0.144.4 or later, installed and signed in. Rill can still be used as a browser without an agent.
Download Rill, move it to Applications, sign in to Claude Code or Codex, and press ⌘E on any page. Rill then looks for the relevant projects in your agent history.
New coding tasks are read-only until you allow edits. In git repositories, edits can be undone. Web tasks stop before payment, ordering, sending, posting, or deleting unless you explicitly tell the agent to continue, and the agent cannot type into password, card, or one-time-code fields.
Rill runs on Macs with Apple silicon and macOS 14 or later. Talking, its voice input feature, is in beta and is enabled from Settings › Talking.
ByteAsk is a terminal-first AI coding agent for C and C++ that edits repositories and verifies changes with the real compiler, debugger, sanitizers, and tests before showing a diff. It offers a free tier plus paid plans, with editor connectors and zero-retention handling described in the source.
MeetStream is a meeting bot API for Zoom, Google Meet, Microsoft Teams, and Webex. It helps developers record, stream, and analyze meetings programmatically, with usage-based pricing and a $5 free credit to start.
CreateOS Sandbox is an isolated compute environment for running code and agent workloads inside Firecracker micro-VMs. It is designed for workflows that need machine-level isolation, private networking between sandboxes, and programmatic control through SDK, CLI, or MCP.
GitBot is a local browser-based workspace for creating reusable coding bots powered by Claude Code, Codex, or OpenCode. It helps developers run those agents across repository folders with persistent threads and configurable permissions.
hob is an independent workspace for coding agents that keeps agent sessions, terminals, history, and follow-up work organized around the tools and providers you already use. It is aimed at developers who want local control over routing, history, and workspace structure rather than a bundled model stack.
Grok 4.7 is an xAI model for software development and knowledge work, including extended coding tasks and professional analysis. It is available through Grok Build, Cursor, the Grok API, and other developer platforms, with token pricing starting at $2 per million input tokens and $6 per million output tokens.