AI Agents
Stories, explainers, and practical ideas tagged AI Agents.
Stories
Your work queue is becoming the next AI interface. Check these four statuses first
New AI agents are turning tasks into work queues. Use four statuses, a realistic example and a reusable prompt to keep human approvals and completed work…
Claude Duet wants to decide what is urgent. Write your alert rules first
Claude Duet is entering research preview. Test a three-level AI alert filter with fictional work messages before connecting accounts or trusting urgency notifications.
ServiceNow wants employees to build AI routines. Start with one meeting brief
ServiceNow announced no-code AI Productivity Routines on October 15. Try a safe pre-meeting briefing workflow with a reusable prompt and five checks before automating.
The next AI security problem may arrive in an ordinary email. Use this five-minute check
New warnings about AI-agent attacks have a practical lesson for anyone using AI with email and documents. Use this five-minute hidden-instruction check before letting an assistant…
Meta’s new Muse Wonder remembers more. Decide what your AI should forget
Meta's Muse Wonder promises AI memory across conversations. Use a practical four-column memory check and correction test to prevent stale preferences and sensitive details from following…
Google’s new Gemini agent can work across apps. Test one handoff first
Google's new Gemini agent can work across business apps. Use this practical handoff test to check decisions, tracker updates and approval boundaries before automating work.
AI agents are starting to sit between you and your apps. Map the workflow before they do
AI agents increasingly act across email, calendars, files and business software. Use this workflow map to decide where an agent genuinely helps.
OpenAI’s Dots can keep working when you are not. Start with the approval map
OpenAI Dots can keep working across apps after you leave the chat. Before giving an always-on agent a recurring job, map exactly what it may read,…
Before you give an AI agent access to your work, run this 3-permission check
AI agents are getting access to more of our work. Use this simple Read → Act → Approve check before connecting one to your files, inbox,…
Microsoft studied 40,000 AI agents. The useful ones are doing very ordinary work
Microsoft analyzed more than 40,000 enterprise AI agents and found that the biggest share are helping with ordinary productivity and support work. That gives teams a…
AI that earns its place in your inbox.
The tools worth trying, a prompt you can steal, and the news that affects your work. No hype, no jargon.