@JohnnyNel_@openclaw Mcp also seemed to be much slower for me
Just some simple python scripts leveraging existing py libs + a cloudflare worker to simplify auth for most normal users (everything is open source on github)
Got tired of everyone asking how to connect @openclaw or Claude Code to Google
`ez-google`. No complex MCPs or bloated setups—just simple CLI tools that any agent can figure out.
Give your AI access to Gmail, Calendar, and Drive and more in seconds
https://t.co/N1UKQ2ZS4F
The amount of crap I get for putting out a hobby project for free is quite something.
People treat this like a multi-million dollar business. Security researchers demanding a bounty.
Heck, I can barely buy a Mac Mini from the Sponsors.
It's supposed to inspire people. And I'm glad it does.
And yes, most non-techies should not install this.
It's not finished, I know about the sharp edges.
Heck, it's not even 3 months old.
And despite rumors otherwise, I sometimes sleep.
Pivoted from Qwen TTS (too heavy for HA)
Pocket TTS just works
• Zero-shot voice cloning, no fine-tuning
• 8 built-in voices
• Fully local, Wyoming protocol
Made a @home_assistant add-on for @kyutai_labs Pocket TTS 🎙️
~100M param model, runs ~10x realtime on CPU. Voice cloning from just a 15 sec sample - no training needed. Mind blown by the results.
https://t.co/JKjkWszyey
Probably due to voice embeddings trained on native language data - "30" maps to 三十 in Dylan's Chinese voice space.
Use Ryan or Aiden for pure English 👍
Been playing with Qwen3-TTS from @Alibaba_Qwen for @home_assistant wyoming tts - it's genuinely impressive! 🔥
But found a hilarious quirk: Using the Beijing dialect voice for English, it says:
"You have a meeting in san shi minutes"
三十 (san shi) = 30 in Chinese 😂
having a monorepo turned out to be a massive advantage for ai coding
all context is inside one repo
api's, servers, auth, landing page, marketing sites, dashboard, ops, everything
@openclaw is utterly cracked.
From my phone, I had to do repo research, indexing all the migrations, edge functions, and tables I have in my supabase
It then passed this context into a Codex agent which used Context7 to pull documentation to help migrate Supabase over to @convex
Codex completed the plan, saved it to my repo, and it's ready for migration.
Keep in mind, this is a task I could have kicked off from the Denny's parking lot.
And to implement the plan? Would have been as instructing it to spin up another Codex (or Claude, Gemini, whatever).
@steipete is a legend. I'm only just scratching the surface, but there's an entire library of skills that I've installed.
Browser automation, remind me, deep research.
It understands images. I can leave it voice memos.
I don't know if this is AGI, but its about as close as you can get right now.
Super excited about this launch -- every Claude Code user just got way more context, better instruction following, and the ability to plug in even more tools
Introducing json-render
AI-generated UI. Deterministic output.
1. Define your component catalog
2. AI steams JSON
3. Render interactive UI
Let users prompt dashboards, widgets and apps - safely constrained to components and actions you define
Had Claude Code build a little plugin that visualizes the work Claude Code is doing as agents working in an office, with agents doing work and passing information to each other. New subagents are hired, they acquire skills, and they turn in completed work. Fun start.
the AI bubble will accidentally trigger a human renaissance
here's what will happen between now and 2030:
phase 1 (2026-2027):
> everyone automates everything
> productivity goes through the roof
> we celebrate never having to think again
phase 2 (2028-2029):
> mass digital burnout
> people realize they've outsourced their cognitive abilities
> depression rates spike because nothing feels earned anymore
phase 3 (2030+):
> the pendulum swings hard in the opposite direction
> outdoor experiences become the new luxury
> manual skills and human connection become premium
> we start valuing struggle and effort again
the faster we rush toward full automation, the faster we'll crave what makes us fundamentally human
that's what i'm hoping for anyway
We're encapsulating all our knowledge of @reactjs & @nextjs frontend optimization into a set of reusable skills for agents. This is a 10+ years of experience from the likes of @shuding, distilled for the benefit of every Ralph
1/ @openclaw is absolutely insane. 🦦 I just built a custom Homeassistant Voice Preview connectetion using nothing but voice notes on Telegram.
Normal intents (lights/fans) stay local & lightning fast on HA. But say "Ask Clawd..." and the real magic starts. 🧵👇
3/ The best part? I didn't write the code. I sent a voice note to Clawdbot, which spawned multiple Claude Code sessions in tmux to iterate on the scripts.
It handled the background logic, debugged the connection, and pinged me when it was live. 🤖🦾