Claude Report — 2026-05-01

🚀 Anthropic Official

Claude for Creative Work — 9 MCP connectors

Launches MCP-based connectors for Blender, Autodesk Fusion, Adobe Creative Cloud (50+ apps), Ableton, Splice, Affinity by Canva, SketchUp, and Resolume Arena/Wire; all nine are available on the free plan. Anthropic also announced partnerships with three art and design educational institutions to integrate the tools into creative curricula.

Claude Security — now in public beta

Opus 4.7-powered vulnerability scanner that traces cross-file data flows and component interactions rather than pattern-matching; generates targeted patch instructions with confidence ratings and severity assessments; exports CSV/Markdown and fires webhooks to Slack, Jira, CrowdStrike, Palo Alto Networks, and Wiz. Currently available to Claude Enterprise customers; Team and Max access coming soon.

API: 1M context window beta retired for Sonnet 4.5 and Sonnet 4

The context-1m-2025-08-07 beta header is now a no-op on these two models; requests exceeding 200k tokens return an error. Migrate to Sonnet 4.6 or Opus 4.6, where 1M context is generally available at standard pricing with no beta header required.

Claude Code v2.1.126 — May 1

Adds claude project purge [path] to wipe all CC state (transcripts, tasks, file history, config) with --dry-run/--interactive/--all flags; model picker now reads from a gateway's /v1/models endpoint when ANTHROPIC_BASE_URL points to an Anthropic-compatible gateway; --dangerously-skip-permissions extended to .claude/, .git/, .vscode/, and shell config files; OAuth login now accepts pasted authorization codes for WSL2, SSH, and container environments that cannot receive localhost callbacks.

claude-api skill lands in CodeRabbit, JetBrains, Resolve AI, and Warp

The open-source skill encoding Claude API best practices — caching placement, compaction, agent patterns, model migration paths — is now bundled in four developer platforms in addition to Claude Code. Available for any coding agent to adopt from the anthropics/skills repository.

🔌 Claude Code Plugins

None of the top 10 updated in window.

Unchanged in window: frontend-design, superpowers, context7, code-review, code-simplifier, github, skill-creator, playwright, feature-dev, claude-md-management.

🛠️ Skills

None of the top 5 updated in window.

Unchanged in window: superpowers:brainstorming, superpowers:tdd, superpowers:systematic-debugging, superpowers:writing-plans, superpowers:executing-plans.

🤖 Agents & Subagents

None of the top 5 updated in window.

Unchanged in window: superpowers:subagent-driven-development, claude-router:orchestrate, ralph-loop, claude-router:fast-executor, claude-mem:do.

🔗 MCPs & Integrations

Meta Ads MCP — official first-party server

Meta ships a beta of its first official Meta Ads MCP server for Claude Desktop and other clients, plus a Meta CLI for terminal agents including Claude Code; covers campaign data reads and ad management operations.

Unchanged in window: modelcontextprotocol/servers, github-mcp-server, filesystem MCP, claude-mem, context7.

💡 Community — Workflows & Ideas

Lessons from building Claude Code: Prompt caching is everything

Anthropic engineering blog distills the core architectural principle: static content (system prompt + tools) must precede dynamic content in every request; switching models mid-session forces a full cache rebuild; compaction forks must replicate the parent's exact system prompt and tool structure to inherit cached prefixes — design for caching from the start, not as an afterthought.

Onboarding Claude Code like a new developer: Lessons from 17 years of development

Developer MacLean documents a "context-as-infrastructure" pattern: a dedicated pwiz-ai repository holds versioned AI context independent of the main codebase; skills are configured with "always load" triggers for specific task types; live MCP servers pull from databases, email, and GitHub simultaneously to give Claude real-time project state.

📰 Quick Mentions

• Claude Code v2.1.123 (Apr 29) — Targeted fix: OAuth 401 retry loop with CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS=1 set.
• Deploying agentic AI across the enterprise with Claude Cowork — Six-month deployment framework guide; Apr 29.
• Product development in the agentic era — Anthropic PM describes building Managed Agents prototypes from sketch to functional agent in a single sitting; Apr 29.
• Building AI agents for the enterprise — L'Oréal, Lyft, and Rakuten case studies on agentic deployment; Apr 30.
• How Kepler built verifiable AI for financial services with Claude — Customer story on compliance-grade verifiable AI in financial workflows; Apr 30.
• Claude service outages — API and claude.ai experienced two incidents: Apr 28 (~6 h elevated errors) and Apr 30 (~30 min down, ~50 min warning).