Release Update

Anthropic Claude September 2026: Build plugins and create work from any conversation

Anthropic Claude September 2026 adds Claude 5.5 models, plugin publishing, conversation-based Cowork work, and artifact creation. It also introduces Salesforce in Claude beta and Enterprise smart reports beta, with plan-specific availability.

By Chris CarpenterAIAI CodingEnterprise AISDLC
Anthropic Claude September 2026: Build plugins and create work from any conversation cover graphic
Back to product release updates
Release facts
VendorAnthropic
VersionSeptember 2026
Published by MeritoOctober 23, 2026
Upgrade pathThe new Claude experience is rolling out gradually to Pro and Max plans on web, desktop, and mobile.
Vendor release notesAnthropic September 2026 release notes
Release historyAnthropic release history

Claude Sonnet 5.5 is now available as a faster, lower-cost complement to Claude Opus 5.5 for teams choosing among the Claude 5.5 models. September also adds new ways to publish plugins, begin larger tasks from a conversation, create artifacts, and inspect Enterprise usage.

Claude 5.5, Fable 5.1, and Mythos 5.1 models

Teams can select Claude Opus 5.5 or Claude Sonnet 5.5 in the new Claude 5.5 family. Anthropic positions Opus 5.5 as performing at the Claude Fable 5.1 level for most work and says it costs 40% less to run than Opus 5. Sonnet 5.5 is positioned as a faster, lower-cost companion to Opus 5.5.

Claude Fable 5.1 and Claude Mythos 5.1 are also now available. Anthropic describes these models as intended for coding and knowledge work. Developers, QA engineers, and AppSec engineers can account for the named model options when they decide which Claude capability to use for their work.

Build and manage Claude directory plugins

Plugin developers can now submit plugins to the Claude directory using a new developer portal. Once submitted, they can follow the review process and, after publication, see usage analytics.

This change gives teams building Claude extensions a defined workflow for directory submission, review tracking, and post-publication usage visibility. It applies to plugin work rather than changing how users access an already-published plugin.

Start Claude Cowork tasks without changing modes

Users can now move from a quick question to a larger Cowork task in any Claude conversation. They can ask Claude to handle work such as a report, spreadsheet, or presentation without first selecting a separate mode. Existing chats, Cowork tasks, projects, connectors, and skills carry into the new experience.

Anthropic is gradually rolling out this experience to Pro and Max plans across web, desktop, and mobile.

Create designs, decks, and documents in conversations

Users can request a design, presentation, or document from any Claude conversation, including Claude Code and the Artifacts tab. Claude Design supports on-canvas editing and importing a design system. Claude Slides provides a dedicated presentation starting point, while Claude Docs supports living documents created with Claude and a team.

Artifacts, including Claude Design, Claude Slides, and Claude Docs, are available on every plan, including Free. For Enterprise plans, these features are beta and remain off until an owner enables them.

Salesforce in Claude beta

Approved organizations can use Salesforce in Claude to bring seller accounts, opportunities, and pipeline information into Claude. The beta plugin includes 37 pre-built sales skills for activities including call preparation, deal review, pipeline dashboard creation, and forecasting.

Salesforce in Claude is available on paid plans only for organizations Salesforce approves through its beta sign-up. The approval requirement and beta status apply to this plugin.

Smart reports beta for Enterprise

Enterprise teams can use smart reports to analyze Claude use across their team. The reports cover work completed, cost, session friction, and repeated patterns that may be candidates for shared skills.

Smart reports are in beta and available on Claude Enterprise plans. The feature is aimed at understanding how work is occurring in Claude, rather than only measuring individual sessions.

What this means for your team

QA, AppSec, and development teams can use the same Claude conversations for quick questions or larger tasks, while artifact tools support shared documents, decks, and designs where available. Teams building Claude plugins can track directory review and usage. Enterprise teams can use smart reports to examine work patterns, friction, and possible shared skills. That shared toolset gives engineering and security leaders a common basis for seeing how Claude work is being performed across teams.

Based on the official Anthropic release notes.

How Merito helps

Merito helps teams adopt and govern AI-assisted delivery practices across engineering, quality, and security work. Explore AI-Powered Software Development and Managed AI Operations for practical operating support.

Merito delivers Anthropic Claude strategy, rollout and operations services, and US public-sector buyers can order Claude licenses through Merito on GSA MAS contract 47QSWA18D008F. Our Anthropic Claude team helps you put the September 2026 changes to work across QA, AppSec, and development, with AI rollout and activation so every role gets the same agentic tooling and AI readiness and governance to keep that access controlled.

About Anthropic products

Merito provides Claude implementation, enablement and governance services. US public sector agencies can also buy Claude through Merito's GSA Multiple Award Schedule contract.

Frequently Asked Questions