Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

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...

1,467,306 görüntüleme • 8 ay önce •via X (Twitter)

34 Yorum

gootecks profil fotoğrafı
gootecks8 ay önce

you're a wizard, WOW and here I was thinking I had gotten some decent mermaid output with a couple of skills put together 🤣 no joke, visual explainer puts the other mermaid skills to shame thank you!!! 🙏

Leon Ho profil fotoğrafı
Leon Ho8 ay önce

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.

Chris Sonnier profil fotoğrafı
Chris Sonnier8 ay önce

@GregThatIsAll This is why I built because I got tired of reading markdown files and there is so much more depth to html files.

Fran Arant profil fotoğrafı
Fran Arant8 ay önce

That looks very cool! I have been having it make text adventure "memory palaces" 😅

OneManSaas profil fotoğrafı
OneManSaas8 ay önce

The CSS pattern library is clutch. I built similar templates for my SaaS docs and it cut explanation time by 60%. Nothing worse than rebuilding the same visual structure every time you need to break down a complex workflow.

neonwatty profil fotoğrafı
neonwatty8 ay önce

At first glance, I thought you were outputting code generation to HTML, but the usage in the readme clarifies usage for tables, charts, etc. Very nice.

Manuel Odendahl profil fotoğrafı
Manuel Odendahl8 ay önce

Been playing with that idea (and similar ideas to the other extension you have with the feedback widgets). Also playing with generating PDF forms to upload to my tablet so I can check things off away from the computer. Didn't get around to cleaning them up since just annotating my documents by hand is currently good enough. Love your stuff.

silv profil fotoğrafı
silv8 ay önce

needed this, thank you

fero profil fotoğrafı
fero8 ay önce

just use mermaid diagrams, no?

Roberto Mello profil fotoğrafı
Roberto Mello8 ay önce

How’s it different from anthropic’s playground plugin?

shpetim profil fotoğrafı
shpetim8 ay önce

This is pretty neat and good effort. But I think this level of "instruction" will overload the AI. Too much wasted on context having to follow those prompts and templates. I prefer to give the AI a loaded gun :)

Bill Wheelock profil fotoğrafı
Bill Wheelock8 ay önce

Yes! This! I've been doing this for awhile, but this is far more formal. A lot of times when discussing a spec or new task with my PM agent, I'll ask for an HTML with dynamic elements to visualize it. Helps if you tell it to add some rizz.

J. Mario Meissner profil fotoğrafı
J. Mario Meissner8 ay önce

This is genuinely incredible, bookmarked 📕 I wouldn’t be surprised if something like this gets integrated into the CLIs themselves soon

Marchen profil fotoğrafı
Marchen8 ay önce

the css pattern library is a nice touch - keeps output consistent without having to specify styling every time

jay profil fotoğrafı
jay8 ay önce

This looks really cool, great job and thanks for sharing 👌🏻

Kelly Sommers profil fotoğrafı
Kelly Sommers8 ay önce

This is fire and thanks so much for sharing. I’m doing something kind of similar but not as good as you Every new feature, optimization investigation, or learning I ask Codex to write me a readme markdowns with high level with links to individual markdown docs with deep dives

Chris Butler profil fotoğrafı
Chris Butler8 ay önce

This is great. I've been doing this on my "explanation" markdown files lately, never got around to writing a skill for it. Will give it a rip!

Vasilije profil fotoğrafı
Vasilije8 ay önce

Rich HTML explanations instead of terminal walls of text. That's a quality of life unlock.

Sheldon ⚡ Hull profil fotoğrafı
Sheldon ⚡ Hull8 ay önce

I've been using anthropic as playground skill and loving it. This reminds me of this but a different feel. I'm definitely gonna give it a try. Nice work.

Tomash Devenishek profil fotoğrafı
Tomash Devenishek8 ay önce

Boss!

KShivendu 🌁 profil fotoğrafı
KShivendu 🌁3 ay önce

Looks promising. I've installed it. Thanks!

Cobalt ☥☥ profil fotoğrafı
Cobalt ☥☥8 ay önce

we converged on making similar tooling within the same ~20 hr period lol. nice

Jesse profil fotoğrafı
Jesse8 ay önce

Brilliant. Thanks 🙏

Maxime Steinhausser profil fotoğrafı
Maxime Steinhausser8 ay önce

Game changer! Thanks 💖 🙏

Josie Kins profil fotoğrafı
Josie Kins5 ay önce

immediately installed and found it much better than the diagrams i've asked codex to produce in the past, thanks!

Shake® profil fotoğrafı
Shake®6 ay önce

this is awesome man, ggz

Julius Gonzalez profil fotoğrafı
Julius Gonzalez8 ay önce

@amorriscode potential core product feature 👀

Saushank.eth profil fotoğrafı
Saushank.eth8 ay önce

@AdamBoudj

Lavine profil fotoğrafı
Lavine2 ay önce

pi-deep-research v0.3.0: pi install npm:pi-deep-research Pair it with @nicopreme's Visual Explainer and the results come back as visual reports instead of a wall of text. The search tool is now a hot-pluggable provider chain. Tavily + Brave built in.

Lautaro.x profil fotoğrafı
Lautaro.x8 ay önce

Very cool

Kelly Sommers profil fotoğrafı
Kelly Sommers8 ay önce

Then I tell it in every deep dive use any mermaid diagrams it thinks would be helpful to teach me topology, interactions, concurrency and sequences. So it’ll generate me flow diagrams or sequence diagrams etc

Noel Smith profil fotoğrafı
Noel Smith8 ay önce

Cool

Hariharan S profil fotoğrafı
Hariharan S8 ay önce

I built something similar but not as beautiful looking as yours. All mods to the mermaid diagrams are so nice. I might steal some.

darshil profil fotoğrafı
darshil8 ay önce

just use obsidian

Benzer Videolar

Introducing /visual-plan - a skill to generate rich, visual plans for Claude Code and Codex. Plan mode in Claude Code is incredible. But I always find my eyes glazing over when it gives me this huge markdown essay in my terminal. I found I can make much better visual plans with reusable components. So I made a skill called `/visual-plan`. It generates plans as MDX with visual, interactive components. Diagrams, interactive API specs, schema design changes, annotated code, and even pan and zoomable wireframes. So for any UI work, you can look at a wireframe first, comment on it, iterate, and then have the agent work. I’ve found this to be a much more intuitive interface for reasoning about what the agent is doing. It’s somewhat inspired by that popular post about how HTML is better than Markdown. But HTML can be slow and verbose to write. And it doesn’t look good checked into a repo. This has really made me feel like humans and engineering are entering a new abstraction phase, where we reason about things at the plan level. As long as the plan is good, agents are getting more and more reliable at executing on it. Almost to the degree that we trust the C compiler to compile to assembly reliably. Plans are the new intermediate representation. I also made a skill for the reverse of this, called `/visual-recap`. After the agent works, it gives you a recap of everything it did. Same idea: wireframes, interactive API specs and diffs, schemas, annotated code, etc. So now when you’re reviewing what the agent did for you, or looking at a pull request of somebody else’s code, you can see a visual recap instead of just reading a wall of text. It’s all free and open source. You can find it on my GitHub. Will link to it in the reply because we all know how dumb these algorithms are with links.

Steve (Builder.io)

126,758 görüntüleme • 4 ay önce

AG-UI makes building agentic applications dramatically easier. Here's how it works. This is a model for a simple chatbot: User → LLM → Response But interactive agents that render UI, pause for approvals, and ask users for input need a much more complex model. When building these agents, a response from the LLM will include a series of state changes as the agent runs: • Agent started a task • Agent called a tool • Agent updated its state • Agent streams these tokens • Agent is waiting on a human • Agent is resuming the task The Agent-User Interaction Protocol (AG-UI) treats the LLM response as a stream of events rather than a text endpoint. In practice, here is what you get as an agent runs: 1. Lifecycle events so your UI knows where the agent is. 2. Text messages that stream tokens. 3. Tool calls so your UI can prefill a form with any required arguments. 4. State updates that keep your UI in sync with the agent. 5. Special events for human approvals, rich media, and custom needs. All of these events travel over standard transports (SSE, WebSockets, or plain HTTP) as JSON. As a result, you can build a frontend that stays in sync with the agent's progress without having to invent a custom process to make this happen. For example, building a human-in-the-loop workflow becomes an off-the-shelf component you can integrate rather than build from scratch. CopilotKit🪁 is the creator of AG-UI, and you can use it when building frontend applications pretty much anywhere: • React • Angular • Vue • React Native • Slack • Teams • Discord • WhatsApp • Telegram Here is the link for you to check it out: Thanks to the CopilotKit team for partnering with me on this post.

Santiago

17,438 görüntüleme • 4 ay önce