Claude Updates & Release Notes

Claude Tag lets you add Claude to a Slack channel, where it works alongside your team. Claude responds when you @-mention it, or proactively when it thinks it can be helpful.

Before, Claude only saw one message at a time, so it made decisions to act proactively based on what was in front of it, but not the wider context of what was around it.

Now, Claude uses context from across the channel, as well as its memory and the standing instructions you have given it, to determine when to contribute to the conversation.

As a result, Claude is now roughly 30% better at determining when, and when not to, proactively respond.

This update comes at no additional cost today. While holding more context does increase Claude Tag’s usage, the additional context Claude Tag holds does not count toward usage or spend limits on any plan.

From passive responder to active participant

Previously a lightweight classifier decided when Claude should act. It looked at each new message on its own and made one yes-or-no call.

For example, here are two engineers chasing the same bug from opposite ends. Neither has a free hour to run it down, and neither message asks for anything.

Priya has a theory. Devon has the evidence. Neither message is for Claude, and neither asks for anything.

Read one at a time, neither message is for Claude, so the classifier correctly does nothing, twice. Read together, there's an obvious piece of work sitting there. One engineer has a theory, the other has the evidence for it, and nobody has time to check.

With the classifier removed, Claude uses context across the channel to make one of four moves:

• Reply inline, when the answer is short, verifiable, and something the channel doesn't already know.
• Start deeper work in a thread, when a message deserves real time.
• Route the message to work it has in flight, when it adds to a workstream Claude already has open.
• Say nothing, when nothing is called for.

Here's the same conversation with Claude Tag using additional context. Claude picks the second move, even without being @-mentioned. It sees Priya's hypothesis and Devon's evidence, opens a thread with the investigation already running, and pulls both engineers in. It acts within the boundaries of the permissions, tools, and scope you have configured.

Same thread, two minutes later. Claude reads the two messages together and starts the work. No @-mention.

The conversations aren't walled off from each other. So when Devon posts an update, it lands in the right workstream. When two investigations turn out to be the same bug, that connection gets made.

Claude now looks at all messages to understand the full context of the channel, to more accurately determine if it should participate in a conversation unprompted.

How Claude decides when not to speak

An annoying agent is worse than an unhelpful one. We built Claude Tag to speak up only when it's useful, and in most channels, on most messages, that means saying nothing.

We do this by grading Claude’s channel-by-channel choices against a rubric based on principles like how useful the comment is, how confident Claude is in the response, and whether there is a person better suited to respond.

Claude also knows when to stop paying attention, similar to how people navigate Slack. It follows a few channels closely while paying less attention to others until someone tags it in. In a channel where, message after message, Claude keeps concluding it has nothing to add, it goes to sleep. A @-mention wakes it instantly.

You can also steer its response behavior in plain language: "Never respond here unless someone tags you," or "Feel free to jump in on anything about the deploy pipeline."

And if you'd rather Claude only spoke in a channel when someone tags it, any member can switch ‘Respond automatically’ off.

The first reply is faster

The additional context also allows Claude to respond more quickly. It acknowledges you in seconds instead of operating silently while it starts up. The work itself takes as long as it always did; what's gone is the silent first minute when you couldn't tell whether it heard you.

Live today

This update is now available across Claude Tag, available for Claude Teams and Enterprise customers. You can get started here. Claude now acts as a more effective collaborator, one that can follow the conversation, decide for itself when to act, and when to stay out of the way.

Add Claude to one channel and watch what it adds to your conversations. Learn more about Claude Tag.

Claude in Chrome

The Claude in Chrome side panel is now a Claude Cowork session. Conversations are saved to your history, your skills and connectors work in the browser, and a task you start in a tab can be finished on the Claude desktop, web, and mobile apps. It’s available on Max and Team plans today, and is rolling out to Pro users over the coming weeks.

Claude in Chrome is a browser extension that lets Claude see the page you're on and take actions in it, including clicking links, typing text, navigating between pages, and filling out forms, using your existing logins.

Many of the tools you use every day connect directly to Claude, but others don't, such as internal dashboards, legacy systems, and vendor portals. With Claude in Chrome, Claude can work in these apps through the browser.

Until now, a session in the side panel was separate from those in the Claude apps, so context and conversations didn't carry between them. Now, the side panel runs the same Claude Cowork session you use on desktop, web, and mobile for longer, multi-step work. Because sessions live with your account rather than a single device, you can start work in a browser and pick it up later somewhere else.

As an example, say you're putting together a budget spreadsheet and need to pull in invoices from several vendor portals. Now, you can ask Claude in Chrome to collect the amounts and dates, and it will open the tabs, read each invoice, and build the spreadsheet. Then, you can pick the session up in the desktop app to add files from your computer, or import last month's budget and ask what's changed, allowing you to maintain context across surfaces as you work.

Understanding the risks

Claude in Chrome carries the same risks as any AI agent that acts in a browser, chiefly prompt injection. Malicious actors hide instructions in web content, such as a web page, an email, or a document. These instructions may not be visible to you, but they can redirect Claude to take actions you never intended.

Since the pilot, we’ve added a check on Claude’s own actions. Use “automatically approve” and Claude works through a task without stopping for permission at every step. Before anything consequential, like submitting a form, sending a message, or downloading a file, a separate check reviews the action against what you originally asked for and blocks anything that doesn’t match. That creates fewer interruptions while maintaining oversight.

Claude still asks before certain irreversible or costly actions, like making a purchase or sharing personal data. While these measures meaningfully reduce the risk, they cannot eliminate it. Prompt injection is a moving target, so we keep hunting for new attacks and building what we learn into each model we release. We recommend starting on sites you trust, and our safety guide has more best practices.

Getting started

To start using Claude in Chrome, install it from the Chrome Web Store, sign in, and open the side panel. The new side panel is available on Max and Team plans today, and is rolling out to Pro users over the coming weeks. On Enterprise plans, Claude in Chrome is off by default. Admins can turn it on and limit it to approved domains. See the admin setup guide.

You’ll still need to use the Claude desktop app to work with files on your computer or with other applications. Claude in Chrome doesn’t run on other Chromium browsers or on mobile yet.

How it works

Claude's Compliance API now covers Cowork across the desktop app, web, and mobile, as well as Claude Code in the CLI and desktop app. Coverage is in beta for Claude Enterprise customers. Compliance and security teams can pull session content and metadata from both products through the same Compliance API interface they already use for Claude chats.

The new endpoints are additive: nothing changes about the data you already pull from the Compliance API today.

Security and compliance teams rely on the Compliance API to see how Claude is used across their organization — for audits and eDiscovery — without deploying separate logging infrastructure for each surface. Extending coverage to Cowork and Claude Code closes a gap: those sessions now show up alongside Claude chats.

The new session endpoints return a consolidated, server-hosted transcript for each Cowork and Claude Code session, so prompts, responses, and tool activity come back together in a single session record.

Each session record carries two kinds of data:

• Session content: prompts and responses, tool calls content (web and MCP), and skills and artifacts content captured as transcript text.
• Session metadata: verified user ID and email address, organization ID, session and per-message IDs, and timestamps.

This beta doesn't include Claude Code on the web, Claude Code accessed through the Claude Platform, or sessions run on Amazon Bedrock, Google Cloud's Vertex AI, or Microsoft Foundry.

Organizations already exporting OpenTelemetry data can keep it running: the Compliance API can work alongside it with no infrastructure required on your side.

Getting started

Coverage for Cowork and Claude Code is available today and included with the Compliance API using your existing Compliance Access Key – there’s no separate integration to build. If it's already enabled for your organization, query the new session endpoints directly. If not, review the Compliance API documentation to enable it.

Now in public beta, self-hosted environments let you run Claude Code sessions on your own infrastructure.

Start a session from the web, mobile, desktop, or a routine, and it runs inside your network, next to your internal services, toolchains, and security controls, rather than on Anthropic-hosted infrastructure.

For most enterprises, we strongly recommend our hosted offering for operational simplicity with no infrastructure to run or maintain. Self-hosted environments are for teams whose network, tooling, or compliance requirements call for keeping agent execution on infrastructure they control. If you go this route, plan to staff engineering to own setup and ongoing maintenance.

Why self-host

We saw organizations in our preview program adopt self-hosted environments for a few key reasons:

• Network access: sessions run inside your network and can reach internal services, databases, and registries without exposing them to the public internet
• Customizability: pre-install compilers, SDKs, and internal CLIs in your environment so every session starts ready to build
• Compliance: source code and build artifacts stay on infrastructure you control

“Self-hosted environments let us integrate Claude Code into our existing development workflows while maintaining our security and operational controls. This setup means Claude can generate PRs, help fix CI issues, and respond to developer workflow events, with compute that can scale based on demand. Claude understands our codebase, making it a strong fit for how our engineering teams build.”
George Jacob, Senior Engineering Manager

Data stays on your infrastructure

Repository checkouts, build artifacts, secrets, and any files a session creates or modifies all stay on infrastructure you provision.

The conversation itself, including prompts, responses, and tool results (which can include code that Claude reads), is sent to Anthropic for inference, and the session transcript is stored so a session can be picked up from any surface.

How it works

When using self-hosted environments, you deploy a set of runners. These long-lived processes pick up sessions and start a Claude Code process for each session. Runners come in two modes.

1. Fixed: you keep a set number running and sessions are distributed across them.
2. On-demand: an orchestrator watches for queued sessions, starts a runner as sessions arrive, and stops them when work finishes so capacity tracks demand.

Runners can serve more than one session, but each session runs in its own checkout, so work stays isolated between developers and accounts. Sessions from every supported surface route to the same environment, so you set it up once and it works wherever your team starts a session.

Note: Self-hosted environments differ from Remote Control, which lets developers continue sessions running on their own machines from a phone or browser. Sessions using Remote Control end when that machine stops running the session and are tied to the user who ran claude, whereas self-hosted environments run sessions on shared infrastructure your platform team operates and can be used by any user.

Getting started

Self-hosted environments are available in public beta to organizations on Claude Team and Enterprise plans. They are off by default and not available for organizations using ZDR.

Plan on a platform, developer experience, or developer productivity team owning setup and ongoing operation, including building and maintaining the runner image, updating runners, and running the orchestrator if you use on-demand mode.

See the documentation to learn more. Share feedback via GitHub or through your Anthropic account team.

How inference hooks works

Inference hooks lets your compliance team inspect and enforce policy on every prompt and tool call response before they reach Claude — across Claude Enterprise surfaces including chat, Claude Code, Claude Cowork, and more. Your DLP server makes the call to block or allow, and Claude enforces that decision in real time, blocking unapproved content before it reaches Claude.

Security teams require every channel where employees can move sensitive data to pass through an inspection point their team controls. Until today, native inline enforcement was limited to Claude Code's client-side hooks. Inference hooks closes the gap with a single enforcement layer that covers every Claude Enterprise surface without separate integration work or agent per product.

When an organization turns on inference hooks, every inference request routes through a signed WebSocket connection to a security server. Before the model starts generating, Claude sends the prompt and its surrounding context to your server. Your server returns a verdict — allow or deny — and Claude only proceeds once it has one. The same check runs on tool calls: when Claude calls a tool — including tools connected through MCP, skills, and plugins — the tool's response is checked before it's sent back to the model.

Ways to use inference hooks

Extend your existing DLP program to Claude. Inference hooks uses an open, webhook-based protocol with a published schema. That makes deployment easy — just point it at the same server your other tools already report to including Netskope, Palo Alto Networks, Proofpoint, Zscaler or an AI security server you built in-house.

Cover chat, Claude Code, Cowork, and additional Claude Enterprise products with one configuration. Turn on inference hooks once at the organization level and it applies to Claude Enterprise surfaces, including tool calls made through MCP connectors, skills, and plugins.

Simplify rollout with shadow mode (always allow), role-based exclusions, and percentage-based rollouts. Customize failure-policy tolerance, timeouts, and other settings to match your organization's risk tolerance.

Getting started

Inference hooks is available today in beta for Claude Enterprise customers. Read the documentation to configure your organization's DLP server and start enforcing policy across Claude Enterprise surfaces.

For security vendors, inference hooks is built on a webhook-based protocol with a documented schema, so you can build an integration, and Claude Enterprise customers can point their organization at your platform.

What's new in MCP

MCP recently surpassed 400M monthly SDK downloads, a 4x increase this year, and has become the industry standard for connecting AI agents to applications. MCP 2026-07-28 is one of the most significant spec releases to date:

• Stateless core.

MCP moves from a bidirectional stateful protocol to a request/response model. Servers can now deploy on serverless and edge infrastructure. This simplifies the experience of building MCP servers for Claude and scaling their usage as they grow in adoption.

• Standardized extensions.

MCP Apps and Tasks now ship under a versioned extensions framework, giving developers a formal path to add capabilities like interactive UIs and long-running work without changing the core protocol.

• Auth hardening.

Authorization now aligns with production OAuth 2.0 and OIDC deployments, so MCP servers connect to enterprise identity systems like Entra or Okta without workarounds.

Companies across the ecosystem have been building on the new spec alongside the MCP community since beta:

“More builders are using our MCP server to bring generated outputs into Figma's canvas, where they can explore, riff and refine them with their team into products that stand out. As that usage grows, our stateless architecture can scale with it, and with MCP Apps, Tasks, and Enterprise-Managed Auth, we can do even more to keep design and code together in one, connected flow.”
Josh Clemm, VP of Engineering

"MCP is the industry standard for connecting AI agents to tools and data, and Intuit is proud to support the new MCP 2026-07-28 spec. The stateless protocol core and extensions framework, including MCP Apps and Tasks, let our technologists and customers build and connect agentic experiences at enterprise scale, and allow Intuit to continue delivering trusted financial intelligence experiences to its 100 million consumers and businesses, wherever they choose to work."
Chris Kasten, Chief Architect and SVP of Engineering, Platform and Development Xceleration Group

"The stateless core in the 2026-07-28 spec makes MCP a first-class HTTP workload with no session management to work around. Our customers wanted MCPs on Netlify to be as simple as the rest of the platform and this new spec unlocks this at its core. Building MCP Apps into the new extensions framework is a huge step forward for scalability, accessibility, and capability across the whole ecosystem."
Sean Roberts, VP of Applied AI

"Moving MCP to a stateless protocol makes it easier to scale our own service and makes it easier for us to add analytics for our customers' MCP servers. This helps us show people how their MCP tools are being used and what tools are missing that their users would want to use. It's great to see this protocol growing in this direction."
Paul D'Ambra, Product Engineer

"Anthropic pairs frontier models with a developer experience that keeps raising the bar. The stateless core in the open MCP 2026-07-28 spec reduces the complexity we manage, so we can ship more features to our customers, faster and at scale."
Andrew Goodman, VP of AI

"At Zoom, we believe organizational context is what enables AI to deliver meaningful work, which is why we've built MCP servers that securely bring Zoom meeting intelligence into AI platforms like Claude. The new MCP spec makes it far easier to deploy and scale MCP servers on standard HTTP infrastructure — so users get Zoom's meeting intelligence faster and more reliably, right inside the AI workflows they depend on every day."
Ross Mayfield, Head of Product for AI Platform

See the MCP 2026-07-28 release announcement for full details on the new spec.

Advancing MCP in Claude

Claude now lists over 950 MCP servers in the connectors directory, used by millions of people every day. This year we shipped support for new protocol extensions alongside features that make MCP easier to build on and deploy:

MCP Apps let servers render interactive UI directly in the conversation. Users can see what a connector is doing and work with it inline, without switching tabs.

Enterprise-managed auth lets admins provision MCP connectors for their whole organization through their identity provider. Admins authorize a connector once, users inherit access through their existing IdP groups, and it's connected on first login: zero-touch setup for the end user.

Observability for developers building connectors gives published connectors in our directory a dashboard showing how they perform across Claude product surfaces. Developers can use it to track adoption, diagnose errors and latency, and break down usage by product.

MCP tunnels (research preview) connect Claude to MCP servers inside a private network without exposing them to the public internet. Teams can bring internal tools to Claude with no inbound firewall rules, no public endpoints, and no IP allowlisting on the origin.

The stateless core, standardized extensions, and hardened auth in 2026-07-28 will help developers bring more applications to Claude, with a lower-friction, more consistent end-user experience. We'll continue investing in MCP as an open standard alongside the community, and in the Claude features that make MCP more accessible and effective in production.

Getting started

Explore the spec and SDKs to get started. Support is rolling out across Claude products soon. If you’re planning to submit your MCP server to Claude’s connectors directory, you can learn more here.

A sounding board that keeps up

Starting today, voice mode runs on Claude Opus, Claude Sonnet, and Claude Haiku, reaches the tools you’ve connected like Gmail and Slack, and speaks many more languages.

Some problems you can't type your way through. Voice mode is for practicing for an important pitch meeting, deciding between multiple offers, reviewing your own process out loud, or brainstorming new ideas.

After we added hands-free conversation earlier this year, people started using voice mode for more than quick questions. They began working through real business problems across longer sessions, out loud rather than through text. But until now, voice mode only ran on our Claude Haiku model, which we chose for speed. It kept conversations quick, but not always deep.

Claude Opus and Sonnet, models designed for hard problem-solving, are now available in voice mode. You can switch models mid-conversation from the model picker. Voice mode uses the fastest version of whichever model you’ve selected, so the conversation runs smoothly. It defaults to the last model you used in text chat, so you can move between voice and text without starting over.

With more capable models, you can talk through a half-formed idea and work out what you actually think. Claude asks follow-up questions and builds on your thinking rather than handing you an answer. Voice mode takes turns, meaning Claude listens, pauses to think, and then responds.

A few things to try:

• Practice for an important conversation and ask Claude for feedback on your communication style.
• Talk through a client pitch and ask Claude to find gaps in your logic.
• Brainstorm ideas for your product roadmap and ask Claude to validate them with competitive research.
• Ask Claude to help you evaluate a few different hypotheses for why your last video went viral.

From talking to doing

When you’ve decided what to do, ask Claude to do it. Some examples of what you can achieve in your tools:

• Running late? Ask Claude to push a meeting on your Google Calendar by 30 minutes.
• Turn a conversation about a client pitch into a one-pager in Canva.
• Ask Claude to summarize the emails you’ve received today and draft responses to the most critical ones.

Claude will ask for permission before using one of your connected tools. You can connect a new tool in Settings > Connectors on the Claude mobile, desktop or web apps.

Converse in many more languages

With many more languages now available in voice mode on every plan, you can talk with Claude in the language you think in, switching mid-conversation. Available languages include:

• English
• French
• German
• Hindi
• Indonesian
• Italian
• Japanese
• Korean
• Portuguese (Brazilian)
• Spanish (Latin America)
• Spanish (Spain)

Tell Claude out loud to switch from English to your language of choice, or select your language from the language picker in voice settings.

Claude doesn’t automatically detect your language, so you’ll need to ask out loud or select your language to make the switch from English. You will need to set your language specifically for voice mode, as your previous language settings will not carry over.

Getting started

The latest update to voice mode is available in beta to all chat users on mobile, desktop, and web, but works best from your phone. On the Free plan, you get Claude Haiku, one connected tool, and all available languages. Paid plans can access expanded models and all your connected tools. Voice conversations count toward your regular usage limits.

Try voice mode in the Claude mobile app by tapping the sound wave icon, or download the app to get started.

A new way to reflect with Claude

Your monthly recap is a new feature at Settings > Reflect that shows you the topics you spent time on, your most active day and peak hour, and observations about how you work with Claude. It's in beta on Free, Pro, and Max plans on the web and Claude Desktop, and requires memory to be on. Alongside it, Settings > Time and focus lets you set optional break reminders and quiet hours. For more information, refer to See your monthly recap and Set break reminders and quiet hours.

Write tools for the Microsoft 365 connector

The Microsoft 365 connector now goes beyond search. With write tools enabled, Claude can draft, send, and organize email, manage calendar events, update mailbox settings, and create and update files in OneDrive and SharePoint. Read and search tools work as before, and Teams remains read-only. Before members can use write tools, a Microsoft Entra administrator needs to consent to the updated permission set and an admin needs to enable them for the organization. For more information, see Set up the Microsoft 365 connector and Connect to Microsoft 365.

Claude Cowork on web and mobile

Claude Cowork is now available on web and mobile in addition to desktop. We are rolling this capability out over the next several weeks starting with the Max plan, with more plans to follow. Cowork runs your sessions remotely (in beta), so your sessions and files are saved to your Claude account and go where you go, on any device. Work continues when you close your laptop, and scheduled tasks run with no device online. Chat and Cowork also share one home now, with one place for your projects and artifacts across both. For more information, see Use Claude Cowork on web, desktop, and mobile.

Write tools for the Microsoft 365 connector

The Microsoft 365 connector now goes beyond search. With write tools enabled, Claude can draft, send, and organize email, manage calendar events, update mailbox settings, and create and update files in OneDrive and SharePoint. Read and search tools work as before, and Teams remains read-only. Before members can use write tools, a Microsoft Entra administrator needs to consent to the updated permission set and an admin needs to enable them for the organization. For more information, see Set up the Microsoft 365 connector and Connect to Microsoft 365.

Claude Code and Claude Cowork are now available in public beta in Claude for Government Desktop, built on the same application our commercial customers use and delivered through a FedRAMP High authorized environment.

With Claude Code, public sector teams can build and modernize the software systems that underpin public services. Claude Cowork works directly with files on the desktop, allowing agency staff to delegate memo creation, RFP reviews, casework, and decks to Claude.

The expanded experience also comes with additional governance capabilities. Administrators can set configuration defaults as well as allocate and control spending across departments. Security teams and authorizing officials get tamper-evident audit logs and documentation that supports the agency ATO process.

Today’s launch makes it easier for agencies to acquire, authorize, and allocate AI in pursuit of their missions.

What's new

Claude Code and Claude Cowork. Agencies get new capabilities on the same cadence as our commercial users. Conversation history is stored locally on the agency-managed device. Inference runs inside a FedRAMP High authorized environment.

Billing that fits appropriations. Program offices can tie AI spend to appropriated funds with standard seats or they can define their own seat tiers with spend and model limits, and usage is purchased in fixed increments with a hard not-to-exceed cap. Administrators can track usage per user and per model in the admin console, and automatic burndown alerts warn them before the balance runs low.

Administration that matches how departments are organized. Department-level administrators can allocate seats and prepaid usage to sub-agencies while allowing each to manage its own users. Administrators can use SCIM group mappings to set rate limits, dollar caps, and allowed models for specific seat tiers. Additionally, layered configuration sets defaults for sub-agencies including what Claude can connect to, which features are available, and instructions that guide how Claude interacts with users.

Oversight by design. Every administrative action is recorded in a hash-chained audit log that organization administrators can review directly in the product. Sensitive operations on Anthropic's side require two-person approval. Usage exports are metering data only so agencies can answer ATO and IG requests without moving sensitive material.

Security and oversight

For security teams evaluating the desktop deployment, we're publishing our FedRAMP Secure Configuration Guide as a public-facing document that customers can use to configure their Claude for Government product in a secure manner.

In addition, FedRAMP requires us to provide our formal change notification, which contains details associated with this change.

Lastly, a penetration-test summary is available for the new desktop client, and subsequent follow up penetration-tests summaries will be provided once available. The change notification and pentest summary are available under NDA through Anthropic’s trust center. The application deploys through standard agency MDM platforms.

Getting started

Claude for Government is available in beta starting today. Anthropic remains the contracted and billing party—agencies don't need a separate cloud-provider relationship to get started.

New customers can request access at claude.com/solutions/government.

Security teams can download the penetration-test artifact through the following link.