Manus AI Just Solved The Biggest AI Agent Problem

Added: 2026-05-07

[00:00:00]Manis AI just solved the biggest AI agent problem. Manis AI just got way smarter with connectors, and this changes how you use AI agents at work.

[00:00:09]Here's what just happened. Manis rolled out a new update called recommended connectors. Plain English, Manis now looks at the task you give it. Then it tells you which apps it needs to finish the job, and it asks you to switch them on with one click. That's the whole thing. The impact is huge. Let me explain why. Up until now, AI agents had a big problem. You'd give them a task, you'd start working, then they get stuck because they couldn't open your Gmail or they couldn't read your Google Drive.

[00:00:34]They couldn't add events to your calendar. So, you'd have to stop, go into settings, find the right connector, turn it on, login, then come back and start the task again. That's dead time.

[00:00:44]That's friction. And for most people, that's the moment they give up and just do the task by hand. Manis just killed that problem. Now, when you give Manis a task, it reads what you're asking. It figures out what apps it needs and it shows you those apps right there in the chat. You click approve. It connects them. The task keeps going. No setup, no hunting through menus. The connectors it can pull in include Gmail, Google Calendar, Google Drive, Notion, GitHub, your browser, and a bunch more. It's custom APIs if you have your own tools.

[00:01:13]Now, here's why this matters for real work. Say you want to plan your week.

[00:01:17]You open Manis. You type, "Look at my emails from this week. find the meetings I need to prep for and put a 30-minute prep block before each one in my calendar. Old way, Manis would say it can't do this without access. You'd go connect Gmail, then connect calendar, then come back and try again. New way, Manis says I need Gmail and Google calendar to do this. Click here to enable. Click runs. That's the whole pitch. It sounds small, but if you've ever tried to use an AI agent for real work, you know setup is where most people quit. Let me give you another example that's actually useful. Say you run a small business. You want Manis to read your last 50 client emails, pull out who hasn't replied in two weeks, draft a follow-up to each one, and save the drafts in your Gmail. The old way, you'd have to set up three or four tools before you even started. Now, Manis just asks you to flip the switches it needs.

[00:02:02]You're going from set up the AI to use the AI. Big difference. Now, here's the part most people are missing. This is also rolling out inside Manis projects.

[00:02:10]A project in Manis is like a saved workspace. You can store your instructions, your reference files, your past chats, all in one place. So every time you start a new task in that project, Manis already knows your style, your tone, your rules. This update, projects can now have recommended connectors built right in. So if you set up a project for, let's say, your client outreach, you can lock in Gmail, calendar, and notion as the recommended tools. Time you or your team uses that project, those apps are ready to go. No setup, no questions. For teams, this is bigger. The admin can recommend connectors for everyone. So when a new team member joins the project, they don't have to guess what tools to connect. Manis tells them, they click.

[00:02:48]That's huge for keeping a team aligned.

[00:02:50]If you want to learn how to use tools like Manis to save time and run more of your business on autopilot, come check out the AI Profit Boardroom. We've got members in there right now using Manis for client work, content, and lead follow-ups. We do four coaching calls every week where you can ask questions about your exact setup. It's daily tutorials walking you through how to set up Manis projects and connectors step by step. And there's a 30-day roadmap built around using AI agents like this in your business. Link in the description or go to apiprofitboardroom.com.

[00:03:20]Okay, back to it. Let me explain why this update is a bigger deal than it looks. Most AI agents fail at one thing.

[00:03:26]They can think. They can't act. They can write you a great email. They can't send it. They can plan your week. They can't put it in your calendar. Connectors fix that. Connectors are the bridge between thinking and doing. They let the AI actually touch your apps and move things around. Here's the catch. Setting up connectors used to be the user's job.

[00:03:44]You had to know which ones you needed.

[00:03:46]You had to find them. You had to log in.

[00:03:48]You had to test them. Most people don't want to do that. Most people just want to say, "Do this and have it work."

[00:03:53]That's what this update is about. Manis is now closing the gap between what you say and what gets done. It's removing the setup work and putting it on the AI itself. Think of it like hiring a new assistant. The old way, you'd have to give them a list of every tool they need access to. Then walk them through getting set up in each one. then test it. The new way the assistant looks at the task and says, "I need access to your inbox and your calendar to do this.

[00:04:16]Can I have it?" You say yes. You start kind of relationship with the AI. It feels less like a tool and more like a real worker. Now, let me show you what's actually possible with this. Here's a use case I think is going to blow up.

[00:04:28]Lead follow-up. You're running a small business. You've got people emailing you, asking questions, going quiet, never closing. You don't have time to chase them all. You set up a manners project. recommended connectors, Gmail, Google calendar, and a CRM like notion.

[00:04:41]Your instructions in the project. Read my inbox every morning. Find anyone who's gone quiet for more than seven days. Draft a friendly follow-up. Add a reminder in my calendar to send it. Log everyone in notion with a status. Now, every morning, Manis [clears throat] runs that. You get a list of drafts ready to review. Your calendar has reminders. Your CRM is updated. You spend 10 minutes reviewing instead of 2 hours chasing. That's the kind of thing this update unlocks. Here's another one.

[00:05:06]Content research. You're trying to grow on social media. You want to write a post every day, but finding ideas takes forever. You set up a project. Connects your browser, Google Drive, and Notion.

[00:05:17]Instructions: Every morning, find five trending topics in my niche. Pull the top three quotes from each. Add a one-s sentence summary in notion. Manus does it. You wake up, open the dock, pick the angle that hits hardest, and write the post in 15 minutes. Idea. Less time, more output. Here's a third one that's super practical. Meeting prep. You've got a calendar full of calls. You never have time to prep for them properly. You walk in cold. The call goes okay, but not great. You set up a project.

[00:05:44]Connectors: Gmail, calendar, drive.

[00:05:46]Instructions. 30 minutes before each meeting. Find the most recent emails from the people on the call. Pull any documents we've shared. Write me a five bullet brief on what we last talked about and what to bring up. Manice runs it on autopilot. You get a brief 30 minutes before every meeting. You walk in sharp. People notice. These are the kinds of things that used to be impossible because the setup was too painful. Now it's a few clicks. Let me hit you with something else worth knowing. This update is part of a bigger trend. Every major AI agent company is moving in the same direction. They're trying to make the AI take care of its own setup. They're trying to make the agent feel less like software and more like a person you hired. Claude has been doing this with their connectors and skills. Chat GPT has been doing it with custom GPTs. Manis is doing it with projects and recommended connectors. The companies that figure this out first are going to win because the gap between AI that's cool and AI that's useful is exactly this setup. It's friction. It's how many steps it takes to get from idea to result. Manis just took out a few of those steps. Now, let me be honest about the limits. You still have to approve each connector. The AI doesn't connect things behind your back. That's a good thing for safety, but it does mean you're still in the loop. Also, this only works for the apps Manis supports.

[00:06:53]If you've got some weird custom tool, you'd have to build a custom MCP connector to get it in. That's doable, but it's not zero work. And there's a learning curve to setting up projects well. The first time you build one, you're going to mess up the instructions. You're going to forget some context. Your first results won't be amazing. That's normal. You get better at it the more you use it. But once you've got a project dialed in, it runs forever. That's where the leverage shows up. Here's the part I want you to think about. If you're sitting here thinking, "Cool, but I don't have time to learn another AI tool." I get it.

[00:07:22]Really do. There's a new AI tool every week. It's exhausting. This is different. This isn't a new tool. This is the same tools you already use talking to each other with an AI doing the boring parts in the middle. Your Gmail, your calendar, your drive, your notion, they all stay the same. Manis just sits on top and moves things around for you. You don't have to learn a new system. You just have to tell Manis what you want in plain English and let it do the work. That's a skill. And the people who learn that skill in the next 6 months are going to look at the people who didn't and think, how are they still doing all this by hand? Here's where I think this is going right now.

[00:07:56]recommended connectors is a quality of life upgrade. It makes Manus easier to use. Cool. But the next step is obvious.

[00:08:02]The AI is going to start picking the connectors fully on its own and just running. You'll set the rules once, prove once, and then forget about it.

[00:08:10]That's already starting to happen. Manis has scheduled tasks, so your AI agent can run on a timer, hitting all your connectors, doing the work, and dropping the results in your inbox or your drive.

[00:08:20]You're moving from I use AI to help me work to AI does the work while I sleep.

[00:08:25]And the gap between people doing this and people not doing this is going to get really big really fast. Okay, let me wrap this up with what to do right now.

[00:08:32]If you're new to Manis, go to Manis IM, sign up and play with one task, just one. Pick something boring you do every week. Try to get Manis to do it. Use the new recommended connectors to skip the setup pain. If you're already using Manis, the move is to set up your first project. Pick your most repeatable task.

[00:08:47]Write the instructions. Lock in your connectors. Then run it twice a week and tweak the instructions until the output is exactly what you want. If you run a team, set up shared projects with recommended connectors so your people don't waste time figuring out which tools to connect. That alone will save hours every week. And if you want help doing all of this with a community of business owners who are doing it right now, come join us in the AI Profit Boardroom. Inside, we've got step-by-step tutorials on setting up Manis projects, the exact connector setups our members use to save time on client work, weekly coaching calls where you can ask about your specific Manis build, a 30-day road map for using AI agents to handle your day-to-day work, a prompt library full of Manis prompts that have been tested, and a member map so you can find people near you who are using these tools, too. Link in the description or go to apiprofitboardroom.com.

[00:09:34]And if you want even more, join the I success lab. It's our free community with over 67,000 members. You'll get the full notes from this video, plus over 100 AI use cases like this one, all free. Link in the comments and description. Look, the way to think about this is simple. AI agents are getting better at handling their own setup. The tools you use every day are starting to connect to each other. The boring parts of work are getting automated. The people who pay attention now are going to pull ahead. The people who wait are going to spend the next year wondering how everyone else got so much faster. Manis just made the on-ramp shorter. use it. That's the update.

[00:10:08]That's why it matters.