AgentSky

Any harness, any LLM — cloud-hosted agents on demand.

SaaS

Developer Tools

Artificial Intelligence

Visit WebsiteSee on Product Hunt

Congrats Darren! The channels part is what sells it for me. I correct my coding agent from the phone mid-run all the time, and that flow is usually held together with duct tape. Curious: when a WhatsApp message lands while the agent is mid-task, does it interrupt the run or queue until the current step finishes?

Congratulations on the launch!
I like the idea and after reading comments i like that context is shared between agents and channels as well as you don't pay if the agent is not used. I would like to try it next dayse with with some of my tasks.

Saying state restore is best effort and cannot prevent a race condition is rare on a launch day. You mentioned wanting idempotent agent replies in your own layer. Where do you think that line sits between the infra and the product on top, Darren?

Reaching a long running agent from WhatsApp or iMessage instead of yet another dashboard feels like how normal people would actually use this. Congrats Darren, the 10,000 plus sessions on tycoon.us before launching is reassuring too. What does recovery look like when an agent goes sideways, can I roll back to any earlier snapshot?

Can AgentSky automatically scale when multiple users start interacting with the same agent? how do you handle sudden spikes in agent sessions?

If you had to recommend one harness for someone building their first long-running agent, which one would you suggest, and why?

congrats! can developers choose different LLM models for the same harness? how quickly can you switch models without changing the agent configuration?

I like that agents stay available across WhatsApp, Slack, and the web. Meeting users where they already work makes a lot of sense.

how does pricing work when an agent is parked but still maintaining its state? is there any cost for storage, snapshots or backups while it is inactive?

Congrats on the launch! 👏
How does AgentSky handle failed or partially completed tasks? can an agent resume from its previous state instead of starting over?

what runtime or channel are you most excited to add next? are you prioritizing integrations based on community requests or technical demand?

Congrats on the launch!
What does the developer API expose compared with the CLI? can teams programmatically create, pause, resume and manage agents?

how does AgentSky handle failed or partially completed tasks? can an agent resume from its previous state instead of starting over?

congratulations! can agents communicate with users across multiple channels simultaneously? for example, could an agent start on Slack and continue the same conversation on WhatsApp?

Suspend and resume is the interesting part, and I think it has a blind spot worth designing for now rather than later.

If parking is free and I only pay while the agent is working, then an agent that has quietly stopped working costs me nothing. Which is lovely, right up until a dead agent and a cheap month look identical on the invoice. For long horizon work the bill has historically been the thing that told people something had gone wrong, and this pricing removes that signal on purpose.

From outside, a suspended agent, a finished agent, and one that crashed and never resumed all present the same way. No activity, no cost, nothing in the channel. So what tells the operator which of the three they have? What I would want is for a long horizon agent to declare an expected cadence when it launches, so the platform can say this one should have woken by now and has not, instead of leaving silence to mean whatever the reader assumes it means.

Second one, and it falls out of snapshots combined with messaging channels. If you snapshot state and restore it, what happens when the snapshot was taken partway through an external side effect? An agent restored to a point just before it sent a WhatsApp message cannot tell whether that message went. Redoing it and skipping it are both wrong, and only one of those is visible to the person on the other end, who receives it twice.

Do side effects get recorded outside the snapshot, so a resumed agent knows what already left the building?

congrats! can one agent use different models depending on the task it is performing? for example could it use a cheaper model for simple tasks and a stronger one for complex reasoning?

The cross-platform memory is the part I’m most curious about. If a user starts debugging a problem through Discord, then continues the conversation through email or a support portal later, does the agent maintain the same context and history, or does each channel create an isolated session?

For developer communities, users rarely stay on one platform, so I’m wondering how agents handle context continuity when conversations move across different channels.

The biggest challenge I see with multi-channel agents is maintaining context. If someone asks a technical question in a community forum today and follows up with the same issue in a private chat tomorrow, how does the agent know it’s the same user and the same problem?

Is the conversation history unified at the agent level, or does each platform maintain its own separate memory?

The multi-channel consistency is the part I want to understand better. If a user starts a thread on Telegram and then messages from Slack the next day, does the agent see a single unified conversation or two separate sessions? For developer communities I manage, people switch channels constantly and that is usually where context breaks happen.