Rohan Seth

Claude Managed Agents @AnthropicAI

Opinions my own

[Gabe @gcemaj]

@karbon0x and I were also around for CMA feedback!

Tag us, DM us, or batsignal us!

[Anthropic @AnthropicAI]

New on our Frontier Red Team blog: using Claude to formally verify software at large scale.

Formal verification can prove that code is free of certain bugs, but historically it has only been feasible for small codebases (or required a lot of expert time).

With AI models, however, this is beginning to change. In collaboration with @theoremlabs, we used Claude to prove security properties in over one hundred real-world open-source libraries over the span of a few weeks.

Read our research in full here: https://t.co/zUyx8CjBEk

[Michael Makarov @micmakarov]

My agents sent me 355 hours of reading this week. I can't read that fast.

We're launching Spacesheep (YC F26) to fix that: agents' work lands in a space your team can actually understand and work on together.

Try it: tell any agent "share this to https://t.co/SPpC2Gonh1"

[Anthropic @AnthropicAI]

We’re committing a portion of Anthropic’s shares to the Anthropic Public Benefit Fund, a new wholly owned entity dedicated entirely to the public benefit.

It will start off with 5% of our shares. For every 50% increase in the company’s value over a set baseline, we can contribute up to an extra 2%, subject to our board’s discretion.

In total, this means that roughly 20% of Anthropic’s shares are committed to the public good. Our co-founders have already pledged to donate 80% of their equity, and around half our employees are donating a portion of their equity as part of a donation matching program we’ve run since our founding.

In the post below, we share more details about our new fund.

https://t.co/XzDntVH0vH

[Rohan Seth @karbon0x]

@amorriscode ♥️

[Claude @claudeai]

Six projects from the last few weeks that caught our attention:

A wildfire analysis prototype that maps the fire’s spread over time and tracks hot spots, made with Opus 5.5.

[ClaudeDevs @ClaudeDevs]

We've been working hard on making the Claude Code desktop app great for running lots of sessions side by side.

It's now much quicker to start sessions, easier to split work across a few, and you can always see what each of them is doing. Here’s what's new:

[ClaudeDevs @ClaudeDevs]

Now in public beta, Claude Platform can send OpenTelemetry traces from the Messages API, Claude Managed Agents, and the Claude SDKs to your observability stack.

[Felix Rieseberg @felixrieseberg]

Hi, I'm Felix. I'm an eng lead working on Claude. I love watching people take flight and do things they never thought possible.

We often describe that as "being more ambitious" when using AI, which is tricky advice to follow. I wrote a tutorial:

https://t.co/vNPPs5FWm8

[Michael Cohen @_hi_mc]

Fantastic writeup here on Flows and why we're all so excited about it:
https://t.co/73hucaO8QC

[Rohan Seth @karbon0x]

@gcemaj @_hi_mc Thank you for believing in it!

[Rohan Seth @karbon0x]

@_hi_mc Not possible without you @_hi_mc 🫡. Keeping me disciplined

[Dave Kiss @davekiss]

HOLY. 410 Claude agents found 100 hidden bugs in under 2 minutes

I had an agent plant one-token bugs in 100 of 300 Python functions and hand the file to a single Claude agent with Managed Agents' new dynamic workflows

It wrote its own plan to review every function then verify the suspects and came back in 112 seconds for $1.81

It found all 100 bug plus 10 more that the answer key thought were false alarms but were actually correct!

I don't even know what to make of this

https://t.co/obO1UywE67

[Rohan Seth @karbon0x]

@bibek_poudel_ Yup, but if you have a subscription plan - you get free credits!

https://t.co/kNx0De8f3Y

[ClaudeDevs @ClaudeDevs]

We’re rolling out monthly Claude Platform API credits for Max and Team plans:

Max 5x: $100
Max 20x: $200
Team: up to $500, pooled

Works on any model, including Haiku 5.5, in your code or third-party harnesses.

How it works and full terms: https://t.co/qmWL6xVGwr

[Rohan Seth @karbon0x]

Dynamic workflows in Claude Managed Agents are an extremely powerful multiagent orchestration primitive, and we wanted to put them in the hands of builders on our platform.

Point a swarm of up to 1,000 agents at a task and don't worry about coordinating them. The Claude Platform takes care of it

[ClaudeDevs @ClaudeDevs]

Claude Managed Agents dynamic workflows are now available in public beta. It's a new type of multiagent orchestration, built for your most ambitious workloads.

A lead agent writes a plan that runs across many agents in phases, combining the results at the end.

[Rohan Seth @karbon0x]

@JeremyDanielFox Thank you, Jeremy! ❤️

[Rohan Seth @karbon0x]

Multiagent topologies are great for search problems, and if you squint hard enough, most problems reduce to one:
• Finding bugs can be modeled as a search problem
• Finding security vulnerabilities can also be modeled as a search problem

My favorite use case so far is autoresearch loops: Claude analyzes the transcripts of its past workflows and writes more optimized versions of workflows to hillclimb on evals.

[ClaudeDevs @ClaudeDevs]

ICYMI we've cut the price of Sonnet 5.5 cache reads in half.

In the Claude Platform, it's now $0.10 per million tokens (input is $2, output is $10).

This means Sonnet 5.5 now runs ~20% cheaper on most agentic work.

(API only, no change to Claude Code usage limits.)

[Willie @bigwilliestyle]

Look ma I'm famous!

Nah fr - Claude Managed Agents changed the calculus for us in building https://t.co/asbioVTiZq

Trust me, we tried other options

Infra is still hard and probably not what you should be doing

[Noah Shinn @noahrshinn]

Instinct in group chats

Starting today, early access users can add Instinct to group chats. A new Instinct joins and works for the whole group.

Making plans with friends usually turns into a frustrating back-and-forth over times and places. With Instinct in the group, you can explore options together, agree on a plan and get it done, all in one thread. Your friends don't even need Instinct to join in.

A few things it’s good at:
- Planning a weekend trip with friends, including dates and arrival times
- Coordinating logistics with a roommate
- Getting tickets as soon as they go on sale, then splitting the cost and sending one to everyone
- Running a fantasy league, keeping standings, settling rules disputes, and reminding everyone to set their lineups
- Sorting out carpool schedules - works out who’s driving each day from everyone’s schedules
- Organizing Thanksgiving, keeping track of who’s bringing what and ordering the shopping list

Group chats are new territory for Instinct, and we put a lot of care into getting the permissions right. Here’s how it works:
- Your personal Instinct asks before connecting with the group’s Instinct. You choose which groups to trust and can remove trust at any time.
- The group’s Instinct has no direct access to your personal accounts. Your Instinct checks your permissions before sharing information or taking action.
- If someone new joins the group, all pending replies from your Instinct are held until you grant it permission to share with the expanded group.

Group chats are available to all of our early access users, and will be rolling out to everyone soon. To join early access for this release, ask your Instinct to put you on the list.

Great work with @GajanNagaraj @anuda_w and @saagarwashere