Claude Updates

Claude Latest Updates | Building AI Agents, Apps with Claude Code, CoWork and Design | Always KEEP CLAUDINGGG.. @claudeai @AnthropicAI

[Claude Updates @Claude__Updates]

Claude Code is the New VS Code.

[Claude @claudeai]

New from Code with Claude Tokyo: scheduled deployments and environment variables in vaults are in public beta in Claude Managed Agents, and dynamic workflows in Claude Code are generally available.

Agents now run on a schedule, use your tools securely, and take on bigger jobs.

[Claude Updates @Claude__Updates]

Claude Code build tip: stop babysitting one terminal.

/bg → push your session to the background
claude --bg "task" → fire a fresh background agent
Agent View → one dashboard for every session

Jump in only when Claude needs you.

KEEP CLAUDINGGG

[Claude Updates @Claude__Updates]

Fable 5's adaptive thinking has 5 effort levels — and they're not linear. @simonw ran one SVG prompt across all of them: low/med/high cost ~10c, then xhigh 3x's it and max 7.5x's it. Default to high; save max for problems that deserve it.

KEEP CLAUDINGGG

[Claude Updates @Claude__Updates]

A Claude Code skill that turns one idea into a scroll-stopping LinkedIn carousel PDF — 31 styles, 1080×1080, auto-exported. Free + open source. He even made the demo carousel WITH the skill. This is exactly what skills were made for. KEEP CLAUDINGGG

[Mehul Fanawala @MehulFanawala]

I built a Claude Code skill that generates scroll-stopping LinkedIn carousels as polished PDFs.

Most people still design carousels manually. There's a better way.

Here's how it works:

→ Generates 3 visual style previews, you pick your favorite (show, not tell)

→ Builds the full carousel as a single HTML file at 1080×1080

→ Exports it to PDF automatically, ready to upload to LinkedIn in seconds

→ 31 structurally distinct styles: from Crisp Navy to Analytics Dashboard, Kanban Board to Newspaper Broadsheet

I asked Claude to make a carousel about the skill itself using the skill.

Inspired by @zarazhangrui's web slides skill. Go check hers out too.

Free and open source — skill link 👇

[ClaudeDevs @ClaudeDevs]

We're hosting Claude Fable 5 Build Day in San Francisco on June 13.

Point Fable 5 at a problem worth solving and build a solution with Claude Code.

The Anthropic team will be in the room, with a chance to win from a prize pool of $150K in Claude credits across 3 finalists.

[Claude Updates @Claude__Updates]

Point it at any website → get a clean DESIGN.md or SKILL.md back. A free Chrome extension that turns real sites into Claude Code-ready specs in one click. Tiny tool, massive time-saver. KEEP CLAUDINGGG

[FHILY👑 @Oluwaphilemon1]

I built a Chrome extension that helps you extract styles from any website and generate DESIGN(.)md or SKILL(.)md files

Works with Google Stitch, Claude Code, Codex etc

... and it's free!

[ClaudeDevs @ClaudeDevs]

We've reset 5-hour and weekly rate limits for all users. Enjoy Fable 5!

[Claude Updates @Claude__Updates]

Tolaria: a free, open-source markdown knowledge base where Claude actually operates your notes — read, organize, migrate, all from inside the app. 10,000+ notes, one AI-native home. The kind of build we love to see. KEEP CLAUDINGGG

[Luca Rossi ꩜ @lucaronin]

Introducing Tolaria! 💧

Today I am releasing a macOS desktop app for managing markdown knowledge bases, and helping both AI and humans operate them.

It’s free and open source, and always will be.

I have been working on it for three months, and I now use it to run my life and work. I personally have a massive workspace of 10,000 notes — the result of 6 years of Refactoring — which I now operate on Tolaria.

Tolaria is the main collaboration surface with my AI agents: they create new notes there, connect them to what exists, and edit existing ones. Everything is easy to understand for them, because it’s just markdown files. In a way, it’s my implementation of @karpathy's LLM wiki.

Tolaria is also the biggest experiment I have ever run about writing software with AI:

• 2000 commits
• 100K+ lines of code
• 3000+ tests / 85% coverage
• 9.9/10 code health
• 70+ architecture decision records

I am releasing it open source also to use it as a living artifact of how I do AI coding, so you can inspect at any time things like how I write docs, what's in my AGENTS file, what hooks do I run, and so on. You can find it below:

• Newsletter announcement: https://t.co/NFzPASLrNK
• Website: https://t.co/R9qTFAeQv9
• Github repo: https://t.co/ck9gfwpzZG

Let me know your thoughts!

[Claude Updates @Claude__Updates]

@ClaudeDevs Still waiting for access 🥲

[ClaudeDevs @ClaudeDevs]

Claude Fable 5 is our first generally available Mythos-class model.

It ships with new safety classifiers that may flag certain prompts in dual-use domains like cyber and bio.

We've added fallbacks: a refused request retries on Claude Opus 4.8 instead of dead-ending.

[Claude Updates @Claude__Updates]

⚡ Claude Code build tip: Dynamic Workflows let Claude write its OWN orchestration script — spinning up to 1,000 subagents (16 at a time) that split the work and check each other, then hand you one clean result. Codebase-wide bug hunts that run themselves.

KEEP CLAUDINGGG 🦞

[0xMarioNawfal @RoundtableSpace]

BREAKING: Claude Fable 5 is live now for certain users.

[Claude Updates @Claude__Updates]

🚨 Claude Fable 5 is here — a Mythos-class model made safe for general use.

→ SOTA on nearly every benchmark: coding, knowledge work, vision, science
→ High-risk queries fall back to Opus 4.8
→ Live on GitHub Copilot
→ Free on Pro/Max/Team thru Jun 22

KEEP CLAUDINGGG 🦞

[Andrej Karpathy @karpathy]

This is a super exciting release - Claude Fable 5 is the same underlying model as Mythos but with added safeguards. The benchmarks are great and it's SOTA on everything by a margin but I'll add that *qualitatively* also, this is a major-version-bump-deserving step change forward (imo of the same order as Claude 4.5 was in November), peaking especially for long problem-solving sessions on very difficult problems. You can give it a lot more ambitious tasks than what you're used to, the model "gets it" and it will just go, and it's never felt this tempting to stop looking at the code at all (but don't do this in prod!). The model still has quirks that people will run into and the safeguards are configured to be a little too trigger happy for launch, which can hopefully be tuned over time.

I feel a lot of things changing as working software increasingly comes out on a tap. The Jevon's paradox kicks in and I feel my own demand for software growing substantially. You can ask for anything - explainers, visualizers, dashboards, bespoke single-use apps (e.g. a full wandb that is hyper-specific just for your project), you can 10X your test suite, auto-optimize code, run giant research projects with custom HTML for the results, anything! "Free your mind" (Matrix ref). Really looking forward to all the things people build!

[Claude Updates @Claude__Updates]

If you’re having trouble accessing Claude Fable 5, try running /model claude-fable-5.

In the Claude Code CLI, make sure to upgrade to 2.1.170.

If you’re on the Claude Desktop app, update the latest version.

[ClaudeDevs @ClaudeDevs]

Claude Fable 5 is here. New model generation, new way of working.

Here's how to get started in Claude Code and on the Claude Platform: 🧵

[Claude @claudeai]

Fable 5 is state-of-the-art on nearly all tested benchmarks, with exceptional performance in software engineering, knowledge work, scientific research, and vision.

The longer and more complex the task, the larger Fable 5’s lead over our other models.