Rill logo

Rill

Freemium
訪問

A native Mac browser that lets Claude Code and Codex work from any page while you browse.

Rillとは?

Rill is a native Mac browser that puts Claude Code and Codex agents beside your web pages. From any page, you can send a task in plain language and continue browsing while the agent works in the background. Rill uses your existing Claude Code or Codex setup and can route requests to relevant local projects based on their history.

Alongside agent workflows, Rill provides browser organization and retrieval tools: tabs can be grouped by activity, browsing is recorded in day-based episodes, and pages can be found by describing what you remember. Agents can also answer questions about a page or compare up to six tabs.

Rill is free during its beta. Agent features require a supported Claude Code or Codex CLI installation and sign-in; ordinary browsing does not require either service.

Rillでできること

Tasks from any page

Press ⌘E to send a plain-language request from the page you are viewing. Rill selects a relevant local project from your Claude Code and Codex history and asks when it is unsure.

Background agent work

Agents continue working in their own area while you browse. Rill shows active agents, their models, tool steps, diffs, and completed work.

Project-aware web access

Agents in a project browse in separate tabs with separate website data and without your logins. Web tasks can use your sign-ins but stop before sensitive actions such as payment or posting unless you explicitly allow them.

Browser organization and retrieval

Rill groups tabs by what you were doing, creates day-based history episodes, and can find a page from a description. Projects touched by agents are shown on a map with AI-named areas.

Page questions and tab comparison

Ask about the current page or select up to six tabs to produce a comparison table that quotes each page.

Local data and browser basics

Browsing history remains on the Mac, passwords are stored in the Mac Keychain, and sign-ins can be brought over from Chrome, Arc, or Safari. Tab sorting and day summaries can be disabled in Settings.

利用シーン

“Delegate a coding task while researching”

From an article, documentation page, or other site, send a request to the appropriate local project and keep reading while the agent works. Review the resulting changes, with edits initially read-only and git changes undoable.

“Research across multiple pages”

Ask about the page you are reading or compare up to six tabs in a quoted table when evaluating sources, options, or related information.

“Handle a web task with a checkpoint”

Ask an agent to navigate a site using your existing sign-in. It can work toward an action such as booking or sending, but pauses before payment, ordering, posting, deleting, or other consequential steps unless you clearly tell it to proceed.

“Return to earlier browsing work”

Use day-based history episodes, descriptive page search, grouped tabs, and the project map to recover pages and agent activity from previous browsing sessions.

よくある質問

What does Rill cost?

Rill is free during its beta. Its AI features run through the Claude Code or Codex plan you already have; the source does not specify additional paid plans or usage charges.

Do I need Claude Code or Codex to use Rill?

You need Claude Code or Codex for AI features, but Rill also works as a browser without an agent. Agent features require Claude Code 2.1.261 or later or Codex CLI 0.144.4 or later, installed and signed in.

Can an agent modify code or complete a purchase without asking?

New coding tasks are read-only until you allow edits, and edits in git repositories can be undone. Web tasks stop before payment, ordering, sending, posting, or deleting unless you plainly tell the agent to continue. Agents cannot type into password, card, or one-time-code fields.

How does Rill handle web and project privacy?

History and passwords remain on your Mac, with passwords stored in the Mac Keychain. Project agents browse in tabs with separate website data and without your logins. Web tasks can use your sign-ins within the stated action limits, and page content is treated as untrusted text.

Which Macs support Rill?

Rill runs on Macs with Apple silicon and macOS 14 or later. Browsing works without an agent, while agent features require a supported signed-in Claude Code or Codex installation.

クイック情報

Product type
Native Mac browser with agent workflows
Platform
macOS 14 or later on Apple silicon
Agent support
Claude Code 2.1.261+ or Codex CLI 0.144.4+
Availability
Free beta
Data model
History and passwords stored on your Mac; passwords use the Mac Keychain
Website
rill.love

Rill のトラフィック分析

トラフィックデータは参考情報としてご利用ください。

ドメイン評価
0

Rillの代替品

deadeye logo

deadeye

deepaksinghcs14.github.io

deadeye is a plugin for coding agents that selects a suitable model and effort level for each task, while trimming verbose command output before it enters context. It is built for Claude Code and also supports Codex CLI, Gemini CLI, Cursor, and Windsurf on an experimental basis.

Pushary Isle logo

Pushary Isle

pushary.com

Pushary Isle is a native Mac app that puts AI agent questions and approval requests in the Mac notch or below the menu bar. It helps people review Claude Code, Codex, Cursor, and other connected agent requests at their desk or answer them from a phone.

OpenTrade logo

OpenTrade

opentrade.io

OpenTrade is an open-source trading harness for Claude Code and Codex agents. It provides a repository-based app, agent templates, and scheduling-oriented run management for automated trading workflows.

Zoku logo

Zoku

heyzoku.com

Zoku is a native macOS app for running and coordinating multiple coding-agent CLIs on one canvas. It lets developers spawn, redirect, monitor, and interrupt named agents by voice while using the subscriptions and command-line tools already installed on their Mac.

Edgee logo

Edgee

www.edgee.ai

Edgee is an agent gateway for coding teams that reduces token usage, routes requests across models, and provides usage visibility. Its Compression V2 works as a drop-in CLI layer for Claude Code, Codex, OpenCode, Cursor, and other supported agents.

Flocker logo

Flocker

flocker.md

Flocker is an AI agent management platform for creating agent profiles, coordinating multi-agent work, and monitoring activity from a shared dashboard. It is designed for people running multiple agents across tools such as Claude Code, Codex, OpenClaw, OpenCode, and MCP-compatible workflows.