Chasing Next
Explore LessonsAI UpdatesBlogAbout
Log inFor TeamsStart for Free
Chasing Next

Learn AI by doing.

Start for FreeFor TeamsAboutFAQUpdatesBlog
Sign UpSign InSupport
2026 Moonswell Marketing LLC·Terms·Privacy··
All Updates

Anthropic Updates

Everything Anthropic has shipped, in one place.

This Week
AnthropicAnthropicfeature
Link·Aug 13, 2026

Claude Tag Channel-Wide Context

Claude Tag now reads the whole Slack channel before deciding whether to jump in, for Claude Teams and Enterprise.

  • Best for: Slack channels where Claude was answering too often, or staying quiet when it should have answered
  • When to use it: your team wants Claude present in a busy channel without it interrupting every thread
  • Who benefits: Teams and Enterprise admins running Claude in shared Slack channels
AnthropicAnthropicfeature
Link·Aug 13, 2026

Claude Code Desktop Auto-Continue

A checkbox in the Claude Code desktop app that picks your work back up on its own once your usage limit resets.

  • Best for: long Claude Code sessions that keep hitting the usage cap partway through a task
  • When to use it: you run out of limit mid-task and don't want to come back later and restart the work by hand
  • Who benefits: people running Claude Code in the desktop app for long stretches
AnthropicAnthropicfeature
Link·Aug 12, 2026

Claude Cowork in the Chrome Side Panel

The Claude in Chrome side panel now runs a full Claude Cowork session, so a task you start in a browser tab continues in the desktop, web, and mobile apps.

  • Best for: Work that starts as browsing and turns into a task: pulling numbers off a vendor portal, filling out a form, checking an internal dashboard, then finishing it somewhere else.
  • When to use it: You need Claude to act inside a site you are already logged into, instead of copying the page contents into a separate chat.
  • Who benefits: Marketers and ops people who live in web tools that have no direct Claude integration, like internal dashboards and vendor portals.
AnthropicAnthropicapi
Link·Aug 11, 2026

Compliance API for Claude Cowork and Claude Code

Claude Enterprise security teams can now pull full session transcripts from Cowork and Claude Code through the same Compliance API that already covered Claude chats.

  • Best for: security and compliance teams that need a record of what people did inside Cowork and Claude Code, not only what they typed into Claude chat
  • When to use it: your organization has to retain or review AI session activity, and those two products were the gap in that record
  • Who benefits: enterprise security, legal, and compliance staff who own retention and eDiscovery
AnthropicAnthropicpricing
Link·Aug 10, 2026

Claude Sonnet 5 Pricing Made Permanent

The launch price for Sonnet 5, $2 per million input tokens and $10 per million output tokens, stays instead of rising after August 31.

  • Best for: Budgeting high-volume agent and automation work that runs on Sonnet 5 through the API.
  • When to use it: You held off building on Sonnet 5 because the cheap rate was scheduled to expire.
  • Who benefits: Developers and teams running large token volumes through the Sonnet 5 API.
Last Week
AnthropicAnthropicfeature
Link·Aug 7, 2026

Claude Managed Agents Update (Budgets, Regions, Skills, Advisors)

Four changes to Claude Managed Agents: a spend cap per session, control over where the work runs, skills loaded from your own repository, and a stronger model an agent can consult mid-run.

  • Best for: Teams running Claude Managed Agents on real workloads where cost and data location matter
  • When to use it: An agent session can run up an unpredictable bill, or your compliance rules require the work to stay in the US
  • Who benefits: Developers and platform teams operating Managed Agents in production rather than testing them
AnthropicAnthropicfeature
Link·Aug 7, 2026

Cross-Session Messaging in Claude Code

Claude Code sessions can now send each other a summary of what they are working on, so you do not have to re-explain a task in a second session.

  • Best for: Running several Claude Code sessions at once and needing one to hand work off to another
  • When to use it: A long job is running in one session and you want to keep working in another without retyping the context by hand
  • Who benefits: Developers who keep multiple Claude Code sessions open across projects or machines
AnthropicAnthropicmodel
Link·Aug 7, 2026

Claude Fable 5 Biology Safeguards Update

Fable 5 now answers everyday health and biology questions it used to refuse or hand off.

Anthropic
  • Best for: Health, medical, and science questions that Fable 5 was declining to answer
  • When to use it: You hit a refusal on an ordinary health or biology question and had to reword it or switch models
  • Who benefits: Anyone using Fable 5 for health, science, or educational work
AnthropicAnthropicfeature
Link·Aug 7, 2026

Claude Code Auto Mode (Default on Pro, Max, and Team)

Claude Code stops asking you to approve each command and auto-approves the safe ones, for Pro, Max, and Team users.

  • Best for: Long Claude Code runs where stopping to approve each step breaks your focus
  • When to use it: You want to review the finished result of a multi-step task instead of every command along the way
  • Who benefits: Developers and technical marketers running Claude Code on Pro, Max, or Team plans
AnthropicAnthropicfeature
Link·Aug 6, 2026

Self-Hosted Environments for Claude Code (Public Beta)

Claude Code sessions run on your company's own servers instead of Anthropic's, for Team and Enterprise organizations.

  • Best for: Teams whose source code, databases, or internal tools are not allowed to leave their own network
  • When to use it: Compliance rules block sending code to an outside vendor, or Claude needs internal services it cannot reach from the public internet
  • Who benefits: Platform and developer-experience teams at Claude Team and Enterprise organizations