Affiliate disclosure: Some links in this article may be affiliate links. We may earn a commission if you purchase via these links — at no extra cost to you. This does not affect our editorial coverage. Full disclosure.
Anthropic Is Retiring Three Claude Features by June 15 — and Everything Moves to VM-Based Execution
TL;DR
Anthropic notified Claude Team and Enterprise admins this week that three features — legacy Artifacts, custom Styles, and Project file upload search — will be retired by June 15. All three are being replaced by VM-based equivalents that require code execution and file creation to be enabled. It is not just a deprecation notice: it is a signal that Anthropic is consolidating all of Claude's power features onto a single sandboxed execution architecture.
June 15
deadline for legacy Artifacts, Styles, and Project file search retirement
1 setting
change required: enable Code execution and file creation in Organisation Settings
3
features migrating: Artifacts → VM Artifacts, Styles → Skills, Project search → VM file search
Anthropic sent an admin-tier notification this week informing Claude Team and Enterprise customers that three features will be retired by June 15, 2026: legacy Artifacts, custom Styles, and Project file upload search. The replacements — VM-based Artifacts, Skills, and an upgraded file search — are already live, but require a specific settings change to activate.
The action required is the same for all three: go to Organisation Settings and enable *Code execution and file creation*. Without it, users on your org will silently lose access to artifacts, custom styles, and Project file search after June 15.
What is actually changing for each:
Artifacts are migrating from Claude's original generation model to VM-based execution. The new VM-based Artifacts (which include Live Artifacts — dashboards that refresh in real time from connected data sources) run in a sandboxed container isolated from your systems. Existing artifacts migrate automatically — no manual action beyond enabling the feature flag.
Styles — the custom writing styles users had saved — are migrating to Skills. A Skill is a lightweight folder containing a SKILL.md instruction file plus optional scripts and assets. Your custom styles become Skills that execute inside the VM environment. The "learning mode" style migrates automatically. If you prefer not to use Skills, you can convert your styles to Claude instructions instead; for org-wide consistency, set instructions at the org level.
Project file upload search is being rebuilt. The new implementation uses VM-based code execution to search and reason over project files, replacing the previous vector search approach. More capable, but requires code execution enabled at the org level to function.
Why Anthropic is doing this now. The shift is about architectural consolidation, not feature removal. Anthropic has been building toward a single execution model: a sandboxed VM environment where Claude can run code, create and edit files, search documents, execute Skills, and interact with external tools — all with admin-configurable egress controls. Legacy Artifacts and Styles were built before this architecture existed. Retiring them clears the technical debt and unifies the platform.
What comes next on this architecture. The VM-based execution environment is the foundation for everything Anthropic is shipping in 2026: Live Artifacts that stay connected to live data sources, Agent Skills that can run Python scripts and multi-step workflows, Computer Use capabilities at scale, and MCP-connected tools. Enabling code execution is not just a migration requirement — it is the gate to Claude's most capable features going forward. Orgs that leave it disabled will increasingly fall behind on capability access as Anthropic ships more features that depend on it.
Why It Matters
This is not a feature removal — it is an architectural reset. Anthropic is consolidating every Claude capability that touches execution, files, or tools onto a single VM-based sandbox model. The practical implication for enterprise operators: the permission you have been deferring (enabling code execution) is no longer optional if you want Claude's core productivity features to keep working after June 15. More broadly, this signals the direction of the product: every major Claude capability in 2026 — Live Artifacts, Agent Skills, advanced file reasoning, Computer Use — runs on this VM foundation. Orgs that embrace the architecture get a significantly more capable Claude; orgs that block it will increasingly face a reduced feature set.
Who's Affected
- — Claude Team and Enterprise admins — action required before June 15: enable Code execution and file creation in Organisation Settings. Without it, Artifacts, Styles, and Project file search stop working.
- — Users who rely on custom Styles — your saved styles are migrating to Skills automatically; alternatively, convert them to Claude instructions for a simpler setup that does not require code execution.
- — Teams using Project file upload search — the new VM-based search is more capable but needs the code execution flag enabled at org level; test it before June 15 to validate your workflows.
- — Security and IT teams — Anthropic has added org-level egress controls (on, limited, or off) and a vetted pre-approved domain allowlist for orgs on limited egress. Review these before enabling.
What To Do Now
- 1. Do this today: Log into claude.ai/settings → Organisation → Skills/Code execution. Enable *Code execution and file creation*. Takes 30 seconds. Prevents workflow disruption on June 15.
- 2. If you have blocked code execution for security reasons, review Anthropic's egress controls first. You can enable code execution with egress set to *limited* — traffic is restricted to a pre-approved domain allowlist, not open internet access. This is the right posture for most enterprise environments.
- 3. Convert styles to Skills or instructions before June 15. Skills give you the full VM-powered version with scripting capability; converting to instructions is the zero-friction option that works without any settings change.
- 4. Use this as a forcing function to evaluate Skills. Agent Skills — reusable workflow modules with SKILL.md instructions and optional Python scripts — are one of Claude's most underused enterprise features. The migration is the right moment to audit what your team actually needs as a repeatable workflow.
- 5. Mark your calendar for June 14. Confirm artifacts still render, Project file search still works, and custom styles/Skills are active. Do not discover the gap on June 16.
More on this topic — Best AI Coding Tools 2026
Independent Review
Claude AI
Pricing, pros and cons, real-world verdict — no affiliate spin.
Read the Claude AI reviewThe AI Hustle Playbook Newsletter
Get the curated shortlist.
A playbook of AI tools and strategies for building income streams.
No spam. Unsubscribe anytime.
More from ToolNav News
Anthropic Launches Claude for Small Business — 15 Pre-Built Workflows, 7 Tool Connectors, and a $200M Gates Foundation Partnership
2026-05-14
Anthropic, Blackstone and Goldman Sachs Launch $1.5B Enterprise AI Services Firm
2026-05-05
Claude Opus 4.7 Is Now Generally Available With 3.3x Better Vision and Coding Gains
2026-04-16