llm/9ad11e16-7acb-4923-bb7e-5d14cd36cf3f/topic-14-df8b8f01-8606-4e90-bd6c-e6797ac054b9-input.json
The following is content for you to summarize. Do not respond to the comments—summarize them.
<topic>
Agentic Workflows and Automation # Discussions extend beyond simple coding to fully automated agentic workflows. Users describe setups where agents running on home servers are triggered via mobile to perform tasks, run scans, or manage infrastructure autonomously. This shifts the mobile interaction from writing code to orchestrating agents that perform the heavy lifting, reporting back status via push notifications or chat messages.
</topic>
<comments_about_topic>
1. I'm using the Android terminal and Claude Code to vibecode on the go. Or rather, as a fairly boring father of two, when I'm tied up in the necessary chores of family life - cooking and cleaning. Nothing as complicated as this - just Claude Code and a fairly standard Linux dev term, but it's remarkable.
Over the recent break, across four or five sessions, I wrote a set of prompts around ~500 words in total.
The result was Claude scanning my network for active ports using nmap, fuzzing those ports with cURL, documenting its findings, self-directing web searches for API/SDK docs for my Hue bridge and ancient Samsung TV, then building a set of scripts to control my lighting system and a fully functional HTML+JS remote for my TV.
The most entertaining part was Claude prompting _me_ to pop into the living room and press the button on the Hue bridge so it could fetch an API key.
The most valuable part? The understanding I gained secondary to generative act. I now understand the button on a Hue bridge literally just tells the device to issue a new API key at the next request. I understand how Entertainment mode works, and why. I understand how Samsung SmartThings is mediated via websockets - and just how insecure decade old Samsung TVs are.
Around 500 words to gain all this? I hate to buy into the hype, but it feels inflectional.
2. I read the Readme. So this is all just stuff you can do with Claude's cli interface? It edits files and runs utilities? And it does this with few enough errors that you can be productive by just chatting with it over ssh? Is Claude the only one that can do this?
3. Possibly Codex, but I've only used Claude Code so far.
Worth pointing out I'm not SSHing to a different device. Claude Code installed and running directly in Android terminal on my phone.
I've built ASP.NET Core APIs on-device this way. Install dotnet in the terminal and Claude can write code, build, run unit tests, and even run the API on localhost. Then use `git` and `gh` to commit, push and raise a PR.
4. How did you make sure Claude wasn't doing anything unintended while allowing it to run scripts it wrote on your network?
5. Claude Code recommended a Telegram bot over email for this very workflow. I've configured my basement RPi to use my "spare tokens" during off hours. At 5PM it messages me to ask if I want the agent to work this evening. If there's no task in the queue I can add one then using the bot. There's also a set of commands to check on status etc. I'm working on the next step to make it a more automated and if there's no specific task, it will create it's own.
6. I've been using Claude Code in their iOS app (on a Pro account). I just point it at my GitHub repo, and tell it to work on one of the issues I created. It required very little setup beyond what I did for Claude Code CLI.
7. Email might work, however if you're a Telegram user you could write a bot that runs on your home system that runs the cli commands on your behalf and then sends the output as a response to you. No need to open up any ports on your router.
8. I didn’t know until I read this comment, but this is exactly what I want. A telegram bot with Claude on the other side and GitHub app to check out the code
9. I've replied with this in another comment, but this seems more pertinent ;)
Thats exactly the approach I took with https://github.com/cloud-atlas-ai/miranda , Telegram bot, PR is the human review point, tests + CodeRabbit catch most issues.
Bot intercepts Claude's AskUserQuestion calls via a hook, sends me an inline keyboard, injects my answer back into the session. Claude keeps working, PR still happens—but I can unblock it from my phone in 5 seconds instead of rejecting a PR based on a wrong guess.
10. I have custom scripts I use at home to keep track of various personal data, assisted by an LLM. The idea of using Telegram as a way to have a global, quick, and personal interface from my phone or tablet, is perfect and easy to set up.
Claude is making it easier to have bespoke data and dashboards for anything. We're going to make a lot of them, for all reasons. I've also made apps with Django interfaces, but quick, global interfaces are going to become in demand.
11. I concur, but I also think that Home Assistant could be used as a rock bed to build many of those dashboards easily. They just need to revert the "go all in on UI first configuration" and keep YAML declarations as first-class citizen to let LLMs easily compose dashboards based on user's desires.
12. I've been using Telegram bot to talk to a Claude SDK agent who talks to my Claudes via tmux commands (all running on a DigitalOcean VPS)
13. I have read of people doing remote coding with clause but through having Claude create pull request. The user then looks through the requests, and either approves or sends it back with edits. Seems like a good way to interact with Claude code, especially once one sets up a test suite and those proposed pull requests have proven not to regress.
14. Same approach here. PR is the human review point, tests + CodeRabbit catch most issues -> https://github.com/cloud-atlas-ai/miranda .
The gap I wanted to fill: when Claude is genuinely uncertain ("JWT or sessions?" "Breaking change or not?"), it either guesses wrong or punts to the PR description where you can't easily respond.
Built a Telegram bot that intercepts Claude's AskUserQuestion calls via a hook, sends me an inline keyboard, injects my answer back into the session. Claude keeps working, PR still happens—but I can unblock it from my phone in 5 seconds instead of rejecting a PR based on a wrong guess.
Works in tandem with a bunch of other LLM enhancers I've built, they're linked in the README or that repo
15. This pretty much sounds like my dream vibe coding dashboard - basically a personal Github populated by AI agents I can assign tasks to. Does this exist yet? Or can something like gitea be setup to behave this way?
16. In terms of issue tracking and agentic "developers", with a mobile focus -
You can connect Linear to Cursor's web agent, which makes Linear issues assignable to the agent directly and kicks off Cursor's take on remote coding agent. You can then guide it further via Cursor's web chat.
If Claude Code on iOS supported Linear MCP (as it does on desktop), you can run a similar issue handoff to agent to issue update workflow, albeit without direct issue assignment to the agent "user". Easy to use labels aka tags for agent assignment tracking, as well.
For my hobby projects, I've been using Linear + agentFlavorOfTheMonth quite happily this way. I imagine Github issues, Asana, whatever could be wired up in place of Linear.
17. Gastown, by Steve Yeggs is that, via tmux. It's rather opinionated and still in development, but it's worth a look if that's what you're looking for.
18. Steve Yegge is building awesome things in this space, but I've found them too heavy, started using bd when it was small, but now its trying to do too much IMO, so made a clone, tailored to my use case -> https://github.com/cloud-atlas-ai/ba
19. durch - just starred this repo! Looking forward to testing it out as I learn how to build with multiple agents.
I'm just starting out with building with Claude - after a friend made this post he sent me a Steve Yegge interview ( https://m.youtube.com/watch?v=zuJyJP517Uw ). Absolutely loved it. I come from an electrical/nuclear engineering background - Yegge reminds me of the cool senior engineer who's young at heart and open to change.
20. This is how I do mobile device coding. Android terminal w. git and gh installed and authenticated. Claude manages the feature branching and PR process; I review the PR in the GitHub mobile app.
21. I code from my phone via GitHub and the Claude actions plugin.
22. I've been working on something similar: https://github.com/shepherdjerred/monorepo/tree/main/package...
Essentially you run a server on some machine. Sessions are created in Docker containers, K8s pods, or via Zellij (an app similar to tmux).
You can:
- Directly attach to sessions via Docker attach (built-in via a TUI). You get a normal Claude Code experience, but multiplexed. The switcher/UI shows you the status of Claude and the PR (pushed, merge conflicts, CI status, review status, etc.)
- Manage sessions via a web UI. Connect to Claude Code directly via your browser. You have access to the usual Claude Code terminal or a native chat view.
- Manage sessions via an app. You have access to a native chat view.
It achieves isolation via Git worktrees + a proxy so that containers have access to zero credentials (there aren't even any Claude code creds in the container), which allows you to more safely use bypass all permissions mode.
This works better for me that Claude Code on Web because I have control over the environment Claude is running in. I can give it any Docker image I want, I can have it connect to my local network, etc.
It's still a WIP (the core bits are there, but it's not polished yet), but I'm hoping it provides a friendlier UX with a similar goal for what the OP has in mind.
23. I've been using a similar workflow for the past couple of months.Heavily inspired by Simon Willison’s approach of building micro tools, I’ve started building micro-utilities. I do this mostly while I'm commuting or outside or waiting for something at work.
Instead of just jotting down an idea in a notes app (and it sitting there for eternity), I’ll open up Jules, describe the tool, and have it scaffold the HTML. I have Cloudflare Pages hooked up to the repo—once Jules submits the PR, the preview branch builds automatically and I can verify the result on my phone immediately.
24. I recommend https://happy.engineering/ . It is very easy to set up. I can have an instance in a container which contains my repository and lots of packages/binaries necessary for the work. I can then use the different binaries to run commands in the container. I was able to easily do `ls -la` in the container and email that to myself, all done from my phone. You can also connect it to applescript and whatnot in order to send sms messages, or you can connect to whatsapp. I was able to make it extract the top 5 headlines on hacker news, get the top ideas being discussed in the comments for each submission, and send all of that into my Apple Reminders for me to read on my phone.
No VPN needed.
25. I'm using this setup as well, and I've been as far as writing a small Telegram bot to send input to Claude when it's stopped running via a Stop Hook
https://github.com/PABannier/claude-telegram-bot
26. If you have GitHub copilot you can create github issues and assign them to copilot. All you need is a browser.
27. I use Prompt, Ever Terminal, Whisper, EC2 and Claude Code.
I can build anything with it. Having Claude on top of a terraform repo lets me fully control my infra. Claude is so good at AWS and terraform, and it even found a $3k monthly accidental spend I had running (also sent a refund request to hopefully get some credit back).
Also have a Claude driven CI workflow in GitHub to help keep everything on track.
Having full access to the Claude Code TUI is so much better than the web or iOS interface, plus everything runs on your own setup.
And agree it has replaced doom scrolling / useless new reading.
28. Because, obviously, you should be spending all of your waking time thinking about LLMs, agents, and how you can integrate them into every part of your life. If you have been living properly in the age of impending-AGI, you would have already been desperately seeking more opportunities to interact with these systems. That desperation would have led you to independently discover agents and all the ways you could couple yourself to them even when away from your computer. Are you a parent stuck at home experiencing life with your kids instead of sitting at your desk? Why not escape such a hellscape by whipping out your phone and building a SaaS from your phone while your offspring annoys you with requests for attention and meaningless affection?
---
Really, this whole environment of 'coding from my phone with dozens of agents while I'm doing the laundry' feels like satire of the sorts of things we used to laugh at on Linkedin.
29. https://github.com/dwash96/cecli
It's a fork of Aider but with agent mode, MCP, skills, task manager and more. Very active development team!
30. I am a huge fan of driving agents from my phone, though this is one of the places where I don’t think terminal UIs work. Agents need a web UI for phones.
31. I just have it send me a push notification.
32. I love this! This concept on steroids is one of the main reasons I made https://github.com/knowsuchagency/vibora after trying both happy.engineering and Vibe Kanban for remote coding. There's the claude mobile app, too, but I want to run Claude on my own hardware in a terminal
33. Does anyone have any good advice or resources on a good workflow to do this with web apps? There's some stuff I'd really like to solve, for myself/family, that would require a front and back-end with persistent storage.
I would love to easily be able to set this up easily when a new idea pops into my mind and then have something running (locally or securely in some cloud) within a few hours/days. I wouldn't want to spend a ton of money for this though, nor have a lot of overhead to manage.
Edit: In addition, I'd like some safeguards where I can't have the LLM of choice accidentally delete stuff or do other unintended things on my network.
34. revived an ipad mini 2 (2013), rooted it and ssh-ed in and let claude handle the tailscale setup, terminal emulator selection, and prep work. perfect form factor and can test web apps via browser.
35. >so much of the AI BS hyping is about inventing supposedly unsolved problems. like Google showing me ads to convince me to use Gemini to write a README.
Okay, but how are you going to write your AGENTS.md file??
</comments_about_topic>
Write a concise, engaging paragraph (3-5 sentences) summarizing the key points and perspectives in these comments about the topic. Focus on the most interesting viewpoints. Do not use bullet points—write flowing prose.
Agentic Workflows and Automation # Discussions extend beyond simple coding to fully automated agentic workflows. Users describe setups where agents running on home servers are triggered via mobile to perform tasks, run scans, or manage infrastructure autonomously. This shifts the mobile interaction from writing code to orchestrating agents that perform the heavy lifting, reporting back status via push notifications or chat messages.
35