Creating no-code AI AGENTS that run locally on your laptop | AnythingLLM

Tim Carambat · Beginner ·🤖 AI Agents & Automation ·1y ago
In this video, I am exicited to announce our "Agent Flow Builder" that now comes in **all forms of AnythingLLM**! The idea is that now anyone can make ai agents with no python setup, no code required, and most importantly, locally. Connect to any API, scrape any website, and do a lot more with AnythingLLM Agents. Download AnythingLLM and make your first flow at: https://anythingllm.com/ Star on Github: https://github.com/Mintplex-Labs/anything-llm Learn about the no-code agent flow builder: https://docs.anythingllm.com/agent-flows/overview Install the two agent flows shown here (put in agent-flows plugin folder) https://gist.github.com/timothycarambat/8ef7d25ce46eff29a045cf650428b036 Chapters: 0:00 - AnythingLLM Introduction 0:50 - Why use the AnythingLLM Agent Flow builder 1:49 - State of Agents: Autogen 2:44 - State of Agents: CrewAI 3:16 - State of Agents: Anthropic MCP 3:45 - Let's get started with AnythingLLM Agent Flow 4:36 - What agent flows look like 5:27 - Available agent flow blocks 8:00 - Demo: HackerNews Headline Scraper 10:50 - Demo: Discord webhook AI message agent 13:13 - Bonus: Chaining agent flows and skills seamlessly 16:11 - Closing thoughts and request for feedback 17:00 - Thanks!
Watch on YouTube ↗ (saves to browser)
Sign in to unlock AI tutor explanation · ⚡30

Related AI Lessons

Chapters (13)

AnythingLLM Introduction
0:50 Why use the AnythingLLM Agent Flow builder
1:49 State of Agents: Autogen
2:44 State of Agents: CrewAI
3:16 State of Agents: Anthropic MCP
3:45 Let's get started with AnythingLLM Agent Flow
4:36 What agent flows look like
5:27 Available agent flow blocks
8:00 Demo: HackerNews Headline Scraper
10:50 Demo: Discord webhook AI message agent
13:13 Bonus: Chaining agent flows and skills seamlessly
16:11 Closing thoughts and request for feedback
17:00 Thanks!
Up next
AI that runs your repo in GitHub : New from Microsoft Research #ai #agenticai #github #workflow
Microsoft Research
Watch →