What I Found
After one week with OpenClaw and my AI assistant Kelex, the biggest lesson is simple: local AI assistants are.
I have been running OpenClaw for about a week, and the honest version is this: it feels like a glimpse of where personal AI assistants are heading, but it is still very much the unpolished era.
OpenClaw is basically a harness that lets an AI model connect to things on your own machine, like Messages, calendars, reminders, browser tools, local files, scripts, and services. In my setup, it is running on a Mac mini and I talk to it through iMessage.
What this video covers
- What OpenClaw Is
- Why I Use Guardrails
- Persistent Memory Is Not ChatGPT Memory
- Model Choice Matters
- My Morning Digest
The short version
OpenClaw is basically a harness that lets an AI model connect to things on your own machine, like Messages, calendars, reminders, browser tools, local files, scripts, and services. In my setup, it is running on a Mac mini and I talk to it through iMessage.
Links & Resources
Project
OpenClaw
The official OpenClaw site.
GitHub
OpenClaw Repository
The source repo for the project.
Support
Mac Support
Apple support for Mac setup, macOS, repair, and troubleshooting.
Official
OpenAI
OpenAI's official site.
App
ChatGPT
OpenAI's ChatGPT app.
Official
OpenAI Safety
OpenAI's safety and responsibility overview.
Official
Anthropic Responsible Scaling Policy
Anthropic's public AI safety framework.
Official
Google AI Principles
Google's AI responsibility principles.
App
Claude
Anthropic's Claude app.
Official
Anthropic
Anthropic's official site.
