Show HN: Computer Agents – Agents that work while you s…

Hey HN, Most AI “agents” I’ve tried are basically chatbots with amnesia — they forget everything the moment you close the tab and can’t do anything unless you’re sitting there watching them. I wanted real AI coworkers that just… work. So I built Computer Agen…

Deploy AI agents to the cloud. They research, code and create.Scheduled, persistent, from any device. Your agent gets its own computer in the cloud with persistent memory, scheduled tasks, accessible from any device. Chatbots forget everything when you close the tab. Your coworker gets its own computer in the cloud with persistent memory, scheduled tasks, and access from any device. Your agents run in the cloud, remember everything across sessions, and work across every device you use.

Every agent gets a cloud workspace that persists across sessions. Files, context, and progress stay there when you come back. Your agents run in the cloud, not on your laptop. Get results on Email, Telegram, or from any device. Deep research with citations, web search, and image generation are built in. Schedule a weekly research report and wake up to results. run a deep research operation to find papers with brillant ideas about possible completely new ways to treat cancer, that are not yet widely recognized, because the authors are not famous enough.

Use the image generation tool to create some very good visualizations when crafting the report about the treatment idea you found and include them there. Key Focus Area: Physics-based therapies appear to be an underexplored frontier. I'm finding promising work on mechanical destruction of cancer cells using ultrasound resonance (Oncotripsy) and molecular-scale vibrations triggered by near-infrared light.

Emerging Paradigm: Non-canonical cell death mechanisms offer novel therapeutic targets. Methuosis, alkaliptosis, and oxeiptosis represent pathways that cancer cells haven't evolved resistance to, unlike traditional apoptosis-inducing treatments. Bioelectric Research: Michael Levin's work at Tufts University on bioelectricity and cancer reversion is particularly compelling. The idea that cancer might be a reversible disorder of cellular communication patterns rather than purely genetic mutations opens entirely new treatment possibilities.

Excellent! Now I'll create compelling visualizations to enhance this comprehensive report. I'll generate: Perfect! Now I'll insert these visualizations into the report at strategic locations. I've successfully completed a comprehensive deep research operation and created a fully illustrated report on under-recognized, innovative cancer treatment approaches. I generated 5 high-quality scientific illustrations and integrated them into the report with captions.

The full 34,000+ character report includes detailed mechanisms, key researchers, and a comprehensive citation list with 51 sources. Access your AI employees from any device. Same agents, same workspaces, always in sync. Isolated execution, explicit data controls, and transparent documentation for teams that need to validate risk before adoption. Each agent runs in its own isolated cloud container with strict resource and access boundaries.

Data is protected in transit and at rest, with layered authentication and access controls. You control your threads, files, and environments with deletion workflows and clear retention policies. Business owners, researchers, and marketers deploy agents that run autonomously, producing real results instead of chat replies. A London-based architect uses aiOS agents to generate professional floor plans in minutes.

From concept to client presentation. FireChatbot uses aiOS SDK and Mail integration to build custom Customer Support Chatbots and Email Assistants. Claude Code power, as a cloud API. Multi-agent orchestration, webhook triggers, and real-time streaming with official SDKs for TypeScript and Python. Define agents with custom system prompts, model preferences, and tool configurations in seconds. Each agent gets its own isolated computer environment ready to execute tasks autonomously.

Every agent gets a dedicated workspace where files persist across sessions. Durable storage that survives restarts and scales automatically. Give agents their own computers! Run code in fully isolated containers with automatic dependency management. Your agents can write, execute, and iterate on code without you building Docker orchestration or managing sandboxes. Extend agent capabilities with pre-built skills like deep research, web search, and image generation.

Skills are modular: Enable what you need and your agents automatically know when to use them. Maintain context across unlimited conversation turns with persistent threads. Your agents remember previous instructions, generated files, and execution history for truly continuous workflows. Coordinate multiple agents in workflows. Run steps sequentially, in parallel, conditionally, or map-reduce. Each agent gets its own thread and workspace, and results flow between steps automatically.

Multi-step research that synthesizes information from dozens of sources. Your agents explore topics thoroughly, cross-reference findings, and produce comprehensive reports with citations. Real-time web search powered by multiple providers. Agents can find current information, verify facts, and gather data from across the internet without you integrating Brave, Perplexity, or Google. Connect your agents to Email, Telegram, OneDrive, Google Drive, Notion, GitHub, and more.

All aiOS integrations are available via API, with no separate OAuth flows or webhook handlers to build. Connect any MCP-compatible tool server to extend agent capabilities. Integrate databases, custom APIs, and third-party services through a standardized protocol your agents understand natively. Fire agents automatically from GitHub pushes, Slack messages, or any webhook source. Each trigger gets a unique URL with HMAC signature verification, with no API key sharing with external services.

Run agents on any schedule with built-in cron support. Set up daily reports, weekly analyses, or triggered workflows without building your own job queue or scheduler infrastructure. Small follow-up, short execution, light model usage. Estimate basis: ~8 CT per task.

Summary

This report covers the latest developments in iphone. The information presented highlights key changes and updates that are relevant to those following this topic.

---