Product Launch
anonymous
22 points
21 comments
Posted3 months agoActive3 months ago
Show HN: Clink – Bring your own CLI Agents, Ship instantly
clink.newAI-powered developmentCLI toolsDevOps
Discussion (21 comments)
Showing 21 comments
This is super cool. Love the idea of using the agents I already pay for instead of signing up for yet another subscription. Curious how you're handling sandboxing and dependency isolation between builds? That's usually the hardest part when trying to make build -> preview -> deploy feel seamless for multiple runtimes.
sandboxing is indeed the trickiest part.
We're using Daytona for development environments, which gives us proper isolation out of the box. Each build runs in its own sandbox workspace with full container isolation, so there's no cross-contamination between projects or users.
Interesting, but Daytona appears to be a cloud-only sandboxing system. Yet your copy states: "coding agents run locally"; so unless I'm missing something, this is at least misleading, since the code/sandboxes need a Daytona cloud account.
3 months ago
ah, I think there's some confusion with the open source version! The OSS Claudable runs everything locally on your machine. This new Clink product runs in the cloud with Daytona sandboxes.
sorry for the confusion, we should clarify that better. the 'locally' copy was from the original OSS project description. will update to be clearer about the cloud infrastructure for Clink!
Interesting use case.
I used many vibe coding apps and also cli agents. This could serve as a useful testbed for evaluating different CLI agents. Does the system support switching between multiple agents, or is that not currently implemented?
3 months ago
For now, we support switching between models within the same provider, but we're currently adding the ability to switch providers mid-session. We're working on agent switching similar to Cursor's LLM switch so you can seamlessly move between Claude Code, Codex, GLM, etc. during development.
We're also exploring auto-switching based on task type, our model would automatically select the best agent for each task. For example, using Opus for frontend work and Codex for backend refactoring. The goal is to leverage each agent's strengths without you having to think about it.
Pretty cool! I see you support deploying to Vercel. Any plans to support other hosts?
Ah, Vercel is what we use in the open source Claudable version! For Clink, we currently deploy static/frontend to Cloudflare Pages and full stack apps to Railway, both work great for different use cases.
We're definitely interested in adding more deployment targets. What hosting platforms would you prefer to see? we can prioritize based on what the community needs most
I'm part of a new hosting platform focusing on the European market, would you be open to discuss what you are looking for and see if we can become partners? Obviously no strings attached :) If so, please drop me an email zander@ploi.cloud
Edit: Forgot to add my thought; I think you should be more clear about what Clink does compared to other software like Claude Code. The building demo's to me (a developer/devops person) are cool but my first thought was 'Why would I use this over Claude Code'? And only in this thread I saw the USP for Clink which makes perfect sense for non-techies!
3 months ago
Thanks for reaching out about the hosting platform! really interested in the European market, will definitely drop you an email once we get things more organized on our end.
You make a great point about the messaging. claude code is definitely comfortable for many devs right now. but we're seeing people constantly switch between codex, claude, and now newer agents like droid and glm. we're positioning ourselves as a mixture of agents platform where you can leverage the best tool for each task.
also yeah, CLI agents are amazing for coding but not great for instant previews or deployments. that's exactly the gap we're filling. the positioning will probably evolve as we learn more from users, but that's the core idea: bring your favorite CLI agents, get the missing pieces (preview + deploy) without paying for another full coding platform.
appreciate the feedback on making this clearer upfront!
Other than having a GUI, how is this different from something more established like openCode ?
Good question! while we use CLI agents as the coding backbone, our focus is really on the application building experience: live preview and seamless deployment.
The key difference is that we give you an isolated sandbox container where you can see your app running instantly, then deploy it with one click. so instead of just getting code output, you get a runnable application that you can interact with and ship immediately. openCode is great for pure coding tasks, but we're optimized for the full build-preview-deploy cycle.
3 months ago
Thanks for the clarification, and kudos to you for making it open-source!
How does this differ from existiong app builders? I've seen quite a few similar projects.....
3 months ago
Most app builders use lightweight coding agents and infrastructure. we're leveraging the full power of CLI agents (Claude Code, Codex, etc) as the engine, which means you can build with any frontend stack like react/next/svelte and any backend like fastapi/nestjs/flask/ruby, then deploy everything as a container. this is live in beta now.
also, if you have an existing project, say a python ML model, you can import it and add a frontend to create a complete deployable service. the CLI agents are just way more capable than typical app builder agents, and we're giving them the infrastructure to match that capability.
Really interesting project! I tried it last week and ran into same errors and today I tried it again and it worked.
I'm curious how the Clink desktop app connects to the web app. Is it sending some kind of oauth token so you can use call Claude Code, etc, on behalf of the user?
FWIW, I created an open-source project in this space https://github.com/dyad-sh/dyad/ and it's great to see people trying different approaches :)
Oh I really like your product! you're right about how it works. the desktop app sends oauth tokens to connect to the cloud infrastructure.
dyad looks really cool, love the local-first approach. it's exciting to see so many different takes on this space right now. what are you focusing on with dyad these days?
3 months ago
thanks! we recently added MCP support so we're figuring out how to make it easier to find & add the right MCP tools for your use case :)
I’ve actually used claudable before, and happy that you guys just made an cloud hosted version of it. But how are you handling all those server costs? Since it’s not just a local open source project right now, the fees might be kinda expensive.
3 months ago
thanks for using claudable before! for server costs, the sandbox previews have auto-stop functionality so the actual runtime is pretty short. keeps costs manageable.
for deployments, cloudflare static hosting is basically free, and we only charge a small fee for custom domains since that's where we have actual costs. the beauty of leveraging existing platforms is that we don't have to run expensive infrastructure ourselves. most of the heavy lifting (the actual coding) happens on your CLI agent subscription, not our servers!
3 months ago
We do the same over at https://rtcode.io with "Agent Share"!
Instead of containers and "stacks", we use the web platform with vanilla HTML, CSS and JS (front-/back-end in 1 place)!
Deployments go to Cloudflare!
See examples at https://new.rt.ht – service workers you can edit live in the playground with URL imports of ESMs, get deployed as Cloudflare Workers!