Posted by admin

In this video, I give a full walkthrough of ClawChat, the application I’m building to manage AI agents as if they were a real working team. I show how I use ClawChat to create agents, place multiple agents inside the same chat thread, assign manager/operator roles, connect agents to real tools, and schedule recurring prompts that keep them working even when I’m not there. In the demo, I walk through real examples from my own setup, including AI agents doing backlink outreach, agents connected to Outlook, Exa Search, X, LinkedIn, and my own local apps, Postman Hermes sorting my email inbox, Social Hermes posting on X and LinkedIn, team chats where agents coordinate with each other, scheduled tasks that wake agents up every hour, reports, analytics, and a live visual “agent operations HQ” showing which agents are working. The bigger idea behind ClawChat is simple: AI agents need more than a chat window. They need a communication layer, a marketplace of tools, permissions, task scheduling, reports, monitoring, and a way for humans to stay in control. ClawChat sits above agent harnesses like Hermes and OpenClaw and lets me connect agents to applications, manage them in teams and departments, and supervise the work they are doing. It is still a work in progress, but I’m building it toward a beta release. If you’re interested in early access to ClawChat, leave a comment below and let me know. I’m also happy to answer questions about how it works, how the agents are set up, or what I’m building next. Chapters: 00:00 Intro: What ClawChat is 00:06 Why I started building ClawChat 00:32 Managing multiple agents in one chat thread 01:10 AI agents doing backlink outreach 02:25 Hourly prompts that keep agents moving 04:36 Connecting agents to Outlook 06:24 Connecting agents to Exa Search and local apps 07:32 Postman Hermes sorting my inbox 08:47 Social Hermes posting on X and LinkedIn 11:09 Creating new agents in ClawChat 14:35 Team chats with multiple agents 17:01 Using manager and operator agents together 17:41 Scheduling tasks into agent chats 18:15 Why Telegram was not enough 20:04 Wrap up, reset, and reports 22:23 Analytics for agent activity 23:19 Agent details and markdown documentation 26:05 Departments, teams, and agent structure 27:20 Calendar and activity tracking 28:07 What my agents are actually doing 30:20 Live agent operations HQ 32:12 Installed apps and marketplace overview 32:29 What ClawChat really is 33:31 ClawChat landing page and beta plan 34:27 Early access and questions 34:51 What comes next #AIAgents #ClawChat #AgenticAI #Automation #AIWorkflow #BuildInPublic