Claude Code @anthropicai. prev @ycombinator W20, mit media lab

SF via Toronto
Joined August 2011
Pinned Tweet
Claude Code can now ask you interactive questions when it needs more information or when there are multiple paths forward.
this and a frontier coding agent
5
1
2
61
Additionally, in response to community feedback we brought back output styles because our alternatives are not quite powerful enough yet. Please keep the feedback coming!
6
1
16
Claude Code Weekly Roundup This week we rolled out a new promo for free credits on Claude Code on the web along with beautiful diffs and support for skills. In the CLI we improved fuzzy search, did a slew of bug fixes & improved triggering for interactive questions.
They wouldn't let me put "be bash tool pilled" on the JD, but.. we are looking for people of culture 🧐
2
18
We're building so many exciting things on the Claude API. Skills, memory and context editing are just the start. The way we build AI apps is going to drastically change over the next year, help us show people the way.
im hiring devrel for Claude Developer Platform to shape how the world builds on Claude APIs this will be the most fun job you've ever had if you love shaping API products, creating technical content, and growing developer communities 🚀 job-boards.greenhouse.io/ant…
11
4
165
every time we make software cheaper to make software engineer salaries go up because more industries enter the market competing over the same talent
13
2
123
one of the best ways to make Claude Code a general agent- browserbase's plugin makes it so Claude can actually use your browser (with your cookies) and take actions using language
I've been using Claude Code completely wrong. I gave it a custom skill and Browser CLI tools and letting it do work for me. It can open pages, click buttons, fill in forms all from your authenticated browser. Just published it to the marketplace, install it in 2 commands.
when the restaurant owner is behind the cashier & takes my order I whisper 'founder mode' under my breath
4
1
60
a clear failure mode I'm seeing for pre-2021 startups: 1. distribution advantage isn't strong enough to fend off AI-native competitors 2. founders aren't hands-on enough with product to rebuild for AI
18
2
2
102
Prompt hooks work by sending the hook input to Haiku to evaluate whether the model should continue or not, and if not what provide an explanation for what it should do instead. Read more about prompt hooks here: docs.claude.com/en/docs/clau…
4
2
47
You can now add prompt-based stop hooks to Claude Code. Prompt hooks are great for encouraging Claude to work for longer periods of time, doing clean up work like removing extra files, writing tests or keeping track of what work is being done.
if the government can own 10% of Intel, I feel like a few grocery stores are fine
15
6
3
168
"code generation is all you need" believers eating good today
New on the Anthropic Engineering blog: tips on how to build more efficient agents that handle more tools while using fewer tokens. Code execution with the Model Context Protocol (MCP): anthropic.com/engineering/co…
the bash tool is all you need
If you're a max user you get $1000 worth of credits to try Claude Code on the web! I would try kicking off long running, compute heavy tasks like starting new features or prototypes on the web and then teleport to the CLI to finish it off and test it.
We're giving Pro and Max users free usage credits for Claude Code on the web. Since launching, your feedback has been invaluable for improving Claude Code. We’re temporarily adding free usage so you can more flexibly experiment with the product.
Thariq retweeted
/stickers
Read more on plugins in the SDK here: docs.claude.com/en/api/agent…
1
14
We've added the ability to use plugins in the Claude Agent SDK. This allows you to bring the same extension points from Claude Code like subagents & skills into your agent. For example, this is how you can use our document skills to create docx, ppt and xlsx files
Thariq retweeted
Claude Code SDK is crazy. For the last few months we’ve been slowly collecting all of our company’s data into a single place and now I’m hacking on an agent that can operate over that data.
24
9
2
359
When building Claude Code our philosophy is to ship fast and learn, while at the same time keeping it intuitive and simple to use. Please keep the feedback coming — we love hearing how we can make Claude Code even better.
4
24