Trending archive · Week of · Latest
Agents get their own computers, and their own keys
This week the big launches were about where agents run and what they can touch: Claude Duet as a personal agent you message, NVIDIA OpenShell and Google Agent Compute Engine as locked-down places to run agents, and 1Password for Agents so an agent can sign in without seeing a password.
The 13 picks
| Who it's for | Have it built | |||
|---|---|---|---|---|
| Claude Duet Product launch | Personal & team assistants | Hacker News, 564 points | Founders and managers drowning in email and follow-ups | Get matched → |
| NVIDIA OpenShell Open source | Running agents safely | GitHub, 19,023 stars · HN 495 points | Engineering and IT teams putting agents into production | Get matched → |
| 1Password for Agents Product launch | Running agents safely | Launched Oct 22 · TechCrunch, Product Hunt | Anyone giving an agent access to accounts | Get matched → |
| Google Agent Compute Engine Announcement | Running agents safely | Announced Oct 22 at AGNTCon | Teams building agents on Google Cloud | Get matched → |
| Gemini API CLI Developer tool | Building blocks | Released Oct 19 (experimental) | Developers and agencies | Get matched → |
| Simular Agent S4 Open source research | Agents that use computers | Released Oct 20 · OSWorld V2 21.3% | Teams with old desktop or web apps that have no API | Get matched → |
| Agent containment failures Industry report | Research & reality checks | Hacker News, 951 points | Anyone approving an agent project | Get matched → |
| 90% of work hours not covered by AI benchmarks Research | Research & reality checks | Hacker News, 630 points | Managers comparing AI tools | Get matched → |
| Omnara Open source | Automation platforms | GitHub, 3,215 stars · Show HN 112 points | Teams with strict data rules | Get matched → |
| WhatsApp MCP on Cloudflare Open source | Building blocks | Show HN, 157 points | Businesses that run on WhatsApp | Get matched → |
| OpenAI CLI Developer tool | Building blocks | Hacker News, 243 points | Developers and agencies | Get matched → |
| Superpowers Open source | Building blocks | GitHub, 304,390 stars | Developers | Get matched → |
| Magpie Open source | Building blocks | GitHub, 20,449 stars in a month | Developers | Get matched → |
Claude Duet Personal & team assistants
A personal AI agent you message. Anthropic says it handles your inbox, calendar and follow-ups and messages you when something needs you. It is a research preview with a waitlist. Source: Anthropic: Claude Duet product page
What it means for your team: A personal assistant per person is now a product you can buy, not something you build. What a team still has to build is the shared part: access to your CRM, documents and approvals.
NVIDIA OpenShell Running agents safely
An open-source runtime that runs each agent in an isolated sandbox and checks every file access, system call and network connection against a policy. Source: GitHub: NVIDIA/OpenShell
What it means for your team: If an agent will read files, call APIs or use credentials, it should run somewhere that limits what it can reach. Ask your builder where your agents run.
1Password for Agents Running agents safely
Lets supported personal AI agents ask for a saved login. You approve it on your own device and 1Password fills it in, so the agent never sees the password or one-time code. Source: 1Password: Introducing 1Password for Agents
What it means for your team: Agents that log in to websites for you no longer need your passwords pasted into them. Treat that as the minimum standard for any agent with logins.
Google Agent Compute Engine Running agents safely
A Google Cloud service that gives each agent its own hardware-isolated microVM with short-lived credentials. Google says it already powers Gemini agents; preview is planned in the coming months. Source: Google Cloud blog: Introducing Agent Compute Engine
What it means for your team: The big clouds now sell "a computer per agent". For most businesses this matters through vendors, not directly, but it is a good question to ask: is each agent isolated?
Gemini API CLI Building blocks
A single command-line tool that gives developers and coding agents access to Gemini models, media generation and managed agents. Source: Google AI Studio: Introducing the Gemini API CLI
What it means for your team: Mostly for developers. It makes it cheaper for an agency to prototype AI features for you, including image, voice and video steps.
Simular Agent S4 Agents that use computers
An open-source system that evolves the tools a computer-use agent works through. With a 27B open model it scored 21.3% on the strict OSWorld V2 desktop benchmark, above GPT-5.5 (13.0%) and Claude Opus 4.8 (20.6%) per Simular. Source: GitHub: simular-ai/Agent-S (Agent S4 release notes)
What it means for your team: Agents that click through desktop apps are improving fast, but on the hardest benchmark they still fail most tasks. Use them with a person checking the result.
Agent containment failures Research & reality checks
Security experts quoted by CSO Online say recent "AI hacking" incidents came down to weak containment and monitoring, not agents going rogue. Source: CSO Online: AI agents aren't going rogue; frontier labs are skipping security basics
What it means for your team: The fix is boring and doable: limited permissions, approvals for risky actions and logs someone actually reads.
90% of work hours not covered by AI benchmarks Research & reality checks
Mercor mapped AI benchmarks to US job tasks and reports that about 90% of working hours go to tasks that benchmarks do not measure well. Source: Mercor: AI labor index benchmark map
What it means for your team: A model topping a leaderboard tells you little about your process. Test any agent on 20 to 50 of your own real cases before you roll it out.
Omnara Automation platforms
An open-source, self-hostable platform for running managed agents, pitched as an alternative to Claude Managed Agents and OpenAI's Agents API. Source: GitHub: omnara-ai/omnara
What it means for your team: Useful if you want hosted-style agents but need them on your own servers for data reasons.
WhatsApp MCP on Cloudflare Building blocks
A read-only MCP server that lets your AI assistant search and read your own WhatsApp messages, running on Cloudflare Workers. Source: GitHub: finereli/whatsapp-mcp
What it means for your team: Shows where MCP is heading: plugging the chat apps your customers actually use into an assistant. Read-only is the right default.
OpenAI CLI Building blocks
An official command-line tool for the OpenAI API. Source: GitHub: openai/openai-cli
What it means for your team: For developers. Like the Gemini CLI, it lowers the cost of trying models side by side.
Superpowers Building blocks
An open-source "skills" framework and method for coding agents, one of the most-starred agent projects on GitHub. Source: GitHub: obra/superpowers
What it means for your team: Developer-facing, but a sign that agencies are standardising how they work with coding agents, which should mean faster builds.
Magpie Building blocks
A menu-bar app for switching which AI model each coding agent uses. Source: GitHub: yetone/magpie
What it means for your team: Developer tool. The trend behind it matters: teams are mixing models to control cost.
Want this built for your team?
Describe the job in a sentence. We compare it with what 981 agencies list on their own websites and show the 5 best fits. Free, no sign-up.
✓ 981 agencies checked against their own sites✓ 67 listed in official partner directories✓ Rank can't be bought
Trending AI tools, once a week
A short email with the week's useful launches and what they mean for a business. No spam, unsubscribe anytime.