Leon Ho

@peduarte @_kpman @raycast I'm experiencing the same issue. The setting is correct, and I’ve tried disabling and re-enabling it, but new history still isn’t saved. It only shows the old entries from the v1 migration.

@_kpman @raycast it should do, can you check Clipboard History's settings?

@raycast I installed the new 2.0 beta version of Raycast, but I noticed that the clipboard feature no longer keeps a history. I didn’t uninstall the old version first — I just installed the beta version over it and let it overwrite the old shortcuts.

@trq212 i firmly believe in this, so i developed an oss tool that enables easy viewing of html artifacts - github.com/agentuse/artif…

@trq212 💯 I wrote an oss small tool to facilitate your fantastic concept: github.com/agentuse/artif…

HTML is the new markdown. I've stopped writing markdown files for almost everything and switched to using Claude Code to generate HTML for me. This is why.

@forgebitz It’s time to rotate into claude

they nerfed gpt-5.5

@pragdua Someone is vibe coding in Google

this design language has to go dawg

@RjeyTech It reminds me of the time Microsoft launched Teams, and Slack responded by releasing a one-page newspaper ad to mock it. The rest, as they say, is history.

Whoop is in full blown panic mode. 😭

try it out

@nicopreme Smart move. I noticed the same thing building agent skills - once output format is standardized, you stop reading and start scanning. That's where the cognitive savings actually live, not in the answer itself but in how fast your brain can parse it.

Created an agent skill called “Visual Explainer” + set of complementary slash commands aimed to reduce my cognitive debt so the agent can explain complex things as rich HTML pages. The skill includes reference templates and a CSS pattern library so output stays consistently well-designed. Much easier for me to digest than squinting at walls of terminal text. github.com/nicobailon/vis…

@deedydas 4% of GitHub commits is the stat that matters most. It means developers aren't just trying it - they're shipping with it. The gap between "impressive demo" and "I can't work without this" is where the revenue lives.

Anthropic is at $14B run rate revenue, the fastest growing software business of all time. – Claude Code run rate is $2.5B in <1yr – $0 -> $100M -> $1B -> $14B in 3yrs – $100k+ customers 7x'd last year – $1M+ customers 40x'd last year, 500+ – 4% of Github commits by CC Insane.

@danhockenmaier The missing axis is self-awareness. Most people don't know which quadrant they're in - the slop cannons think they're in the top right. AI amplifies judgment you already have. It can't give you judgment you don't.

Four types of people at every company now yes, people get 10x better when the go from bottom right to top right but also, people get 10x worse when they go from bottom left to top left

the reason for the release of Claude Opus 4.6 is probably to support the agent teams feature. `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1` to enable it.

@fkysly AgentUse github.com/agentuse/agent…

现在大家开发 Agent 用的都是什么框架？ LangChain? Vercel AI SDK? Claude Agent SDK? Pi-mono？ 从零手写？ 还有什么？ 哪家更好一点

@aiktp_com @tom_doerr Check out the example at agentuse.io :) Simply run `npx -y agentuse@latest run agentuse.io/hello.agentuse` to try it out!

Runs AI agents from Markdown files github.com/agentuse/agent…

@heyblake agentuse.io

Fork it Drop your homepage URL I'll give 1 fix to as many as I can

@d4m1n very cool! how is the quality of the output? are you accepting most of its outputs?

Ralph loop is extremely unhealthy. I now run 1-2 loops 24/7, tweaking, iterating. Before sleep I set off a loop, I wake up 3-5x a night thinking of it with excitement. My workflow is getting better, but it's a steep learning curve. I'll share it all when I'm a bit further, there are some things I still don't like. The best thing? I am doing projects I never had time to do and I can't stop. I am tired too 😅 But having something code while you sleep is pretty incredible.

@AAMurphy8 yes, the "vacation mindset" vs. the "fed up" trigger really captures why January 1st resolutions fail.

@leonho Glad I read this post today. Have been mulling over how I love fresh starts yet few fresh starts become permanent changes for the better. Realized I plan for fresh starts as if they are a vacation. Real changes come unscheduled -- I just get fed up, fired up. and then I change.

@trq212 @tenobrus I think eventually we might need skill search at least. The number of installed skills will naturally only go up, not down, as the usage goes up.

So skills, agents, etc. are a bit weird because they are not tools but instead parameters that the Skill and Agent tools can call with It's not clear to me if there should be agent search, skill search, one unified search extension point or no such search. I think what's unique about MCP is that you're not really in control of how much they put into context, whereas subagents and skills put roughly a constant amount of information in context so you can curate yourself a bit more (e.g. if you install 50 skills that's on you).

We're rolling out Tool Search in Claude Code to reduce how much context MCP servers take up.

AgentUse went from v0.1 → v0.5 in 5 months. Here's what we learned building an agent runner that works with any model and runs on autopilot: