Claude gets a separate toolset for driving a browser
August 19 brought two changes to how Claude operates someone else's screen. The computer use tool moved to general availability as computer_toolset_20260801, with no beta header, actions grouped into a single turn, and zoom enabled from the start. Alongside it came browser_toolset_20260801, a separate set of 31 tools that works inside a browser you run yourself rather than across the whole desktop. The important difference is that this set reads the page accessibility tree and returns element references, so a click targets a reference such as ref_2 instead of a point in pixels — and that survives a layout shift.
On August 19, Anthropic split two jobs that had been going through the same tool. Driving the whole desktop stayed in computer_toolset_20260801, which left beta the same day, while working inside a page got its own browser_toolset_20260801. The difference is not cosmetic: the new set sees the structure of a page, not just a picture of it.
Thirty-one tools, sorted by the job
The set ships with 27 tools enabled from the start and four more you turn on yourself. The split follows what people actually do in a browser.
- Navigation and capture: navigate, screenshot, and zoom.
- Pointer: clicks with every button, double and triple click, hover, drag, separate press and release, movement, and scrolling.
- Keyboard and timing: type, key,
hold_key, and wait. - Reading the page:
read_pagereturns the accessibility tree with references, find locates an element from a plain-language description, andget_page_textextracts visible text, favoring the main content. - Forms and tabs:
form_inputsets a field value directly, whilenew_tab,list_tabs,switch_tab, andclose_tabmanage tabs.
An element reference instead of a point in pixels
Actions take a target in one of two forms. The first is a viewport coordinate, x and y, the same shape the computer use tool understands. The second is an element reference such as ref_2, returned by read_page and find.
The second form is what makes this set usable on real sites. A reference survives a layout shift and requires no pixel guessing, so a button that moved after an image loaded is still hit. References hold until the tab navigates to a new address or the page structure changes substantially, and they belong to a single tab.
Browser state comes back with the result
Results can carry a browser_state block, in which the executor reports what actually happened in the browser: which tabs are open, with title and address, which one is active, whether any opened or closed, and whether a download started, finished, or failed.
- A single block holds at most 100 tabs, and exactly one carries the active marker.
- A tab reference is a stable string of at most 4,096 characters, with no control characters.
- Several actions in one turn run in order, but after the first failure the rest do not run and a message says so.
What stays off until you turn it on
Four tools ship disabled, and for good reason. Running JavaScript in the page and uploading files give the most power, while reading the console and network traffic serve diagnostics. You enable them individually through the configs array, and only once your executor actually supports them and the task calls for them.
Where it runs and where it does not
- Supported models are Claude Fable 5, Claude Mythos 5, Claude Opus 5, Claude Sonnet 5, and Claude Opus 4.8.
- The set is available only on the Claude API. It is not on Amazon Bedrock, Google Cloud, or Microsoft Foundry, and not within Managed Agents.
- There is no separate billing. Usage counts as ordinary token consumption, with the caveat that page reads and screenshots consume input context.
- The browser runs in your environment, not on Anthropic's servers.
Security is a condition here, not a recommendation
The documentation is explicit on this point, because executing someone else's content and taking real actions meet here. The browser should run in an isolated environment without credentials, the allowed-domain list belongs on the network, and navigate should accept only http and https addresses. Actions with consequences, such as a purchase or an account change, require a human to confirm.
„Pages can contain prompt injection text.“
Sources
Related
OpenClaw 2026.10.2 stops long tool output from slipping credentials past redaction
A bug in OpenClaw's secret masking let a credential through whenever it straddled the boundary between two slices of a long text. The fix ships in 2026.10.2 along with repaired updates from 2026.10.1 and a new compaction default for OpenAI and Anthropic API keys.
OpenClawverified

OpenClaw 2026.10.1 lets agents search installed skills and changes how sessions message each other
OpenClaw released version 2026.10.1 on October 10, 2026, its first stable release since 2026.9.7. Agents can now search the skills installed on a machine and read only the one a task needs, instead of carrying the whole library in the prompt. Opt-in MCP Apps run beside chat, and full backups can go to an external drive or Cloudflare R2. Three changes can break existing setups: session-to-session messaging, the old required argument, and Workboard data stored before July 2026.
OpenClawverified
Claude Code 2.1.294 fixes prompt hooks that let through the actions they were written to block
Anthropic released Claude Code 2.1.294 on October 8, 2026. It fixes prompt and agent hooks phrased as instructions, which could allow exactly the actions they were meant to stop. Version 2.1.295 followed the same day with an onFailure setting that makes a crashed or timed-out command or HTTP hook block the action. Releases 2.1.292 and 2.1.293 from earlier in the week closed several permission and sandbox gaps.
Claude Code Releases (GitHub)verified