An Van

Building Agent River.
Stop waiting for your Claude Code agents.
Let them wait for you.

[An Van @anvanvan]

1/ I'm building Agent River.

AI coding agents are insanely capable. The bottleneck isn't the AI.

It's you. Your attention. Your context switching.

Working with agents should feel intentional and fun. Not frantic.

[An Van @anvanvan]

@claudeai Seeing an official agent view implementation gives me confidence I‘m onto something with my Agent River project.

Agent River is agent view on steroids and let’s you flow across multiple projects and juggle dozens of agents while protecting your attention health.

[ClaudeDevs @ClaudeDevs]

We fixed a bug where rate limits on Claude subscriptions weren't properly adjusted for long context requests in Opus 4.7.

We've reset 5-hour and weekly rate limits. Enjoy Opus 4.7!

[An Van @anvanvan]

@odd_joel It works great. Thanks for the quick shipping 🙏🏻

[An Van @anvanvan]

@odd_joel Love the 1976 one

[Claude @claudeai]

Introducing Claude Design by Anthropic Labs: make prototypes, slides, and one-pagers by talking to Claude.

Powered by Claude Opus 4.7, our most capable vision model. Available in research preview on the Pro, Max, Team, and Enterprise plans, rolling out throughout the day.

[Boris Cherny @bcherny]

Opus 4.7 uses more thinking tokens, so we've increased rate limits for all subscribers to make up for it. Enjoy!

[An Van @anvanvan]

@dan_note Would be cool to have an CLI for agents

[An Van @anvanvan]

Can't wait to try out the new /ultrareview

[An Van @anvanvan]

Update to Claude Code 2.1.111 to use Opus 4.7 xhigh

[An Van @anvanvan]

Auto Mode is finally available for Max subscribers too

[Adiel Yaakov @Adielyaakov]

PSA: If you can't get Opus 4.7 on claude code CLI, just trick it via this flag: 'claude --model claude-opus-4-7"

It will say its Opus 4 but its model ID is 4-7.

You're welcome

[An Van @anvanvan]

"Rewinding Instead of Correcting"

[Thariq @trq212]

https://t.co/Ljzw0lmvao

[Claude @claudeai]

Introducing Claude Opus 4.7, our most capable Opus model yet.

It handles long-running tasks with more rigor, follows instructions more precisely, and verifies its own outputs before reporting back.

You can hand off your hardest work with less supervision.

[An Van @anvanvan]

@odd_joel The UX of the d-pad is much better now!

Forward delete would also be useful 😉

[An Van @anvanvan]

The highest ROI is still in investing in fundamentals, like expressing yourself. Which will benefit prompting and many other areas of your life.

[Thariq @trq212]

I think "prompting" will keep being an incredibly high-leverage skill, like writing or public speaking.

It is the skill of talking to agents, mediated by the harness.

My main goal is to grow the bandwidth between humans and agents, to help us understand each other better.

[Noah Zweben @noahzweben]

Thrilled to announce the Monitor tool which lets Claude create background scripts that wake the agent up when needed.

Big token saver and great way to move away from polling in the agent loop

Claude can now:
* Follow logs for errors
* Poll PRs via script
* and more!

[An Van @anvanvan]

What's your Claude Code /buddy aka tamagotchi?

[Naval @naval]

Vibe coding is more addictive than any video game ever made (if you know what you want to build).

[An Van @anvanvan]

Use skill chaining to make your skills token efficient by avoiding redundancies

[akira @realmcore_]

https://t.co/0xm0EmlygR